No Causal Relationship Between Helicobacter Pylori Infection and Rosac

Introduction

Helicobacter pylori (H. pylori)¹ is a bacterium known for colonizing the stomach lining and causing gastrointestinal conditions such as chronic gastritis, peptic ulcers, and gastric cancer, has also been investigated for its potential role in extra-gastric diseases,² including rosacea.³ While the exact relationship between H. pylori and rosacea remains under investigation, the evidence suggests that H. pylori may play a role in the pathogenesis or exacerbation of rosacea in certain individuals.⁴ Further research is needed to clarify the mechanisms and establish whether H. pylori eradication should be considered as part of the treatment strategy for rosacea patients with concurrent H. pylori infection.

Previous studies have shown that patients with H. pylori infection may have an increased risk of rosacea,⁵ and the cure of H. pylori led to partial regression of the skin symptoms in rosacea patients.^3,6 However, whether there is a potential link between rosacea and H. pylori infection remains uncertain, and no genetic association studies between the two have been conducted up to now.

GWAS database is a public resource that stores and shares data. By analyzing associations between genomic variants and specific traits or diseases across large populations, GWAS aims to identify genetic links to complex diseases, physiological traits, or drug responses. Mendelian randomization (MR) is an epidemiological approach that employs genetic variants as instrumental variables to proxy for the exposure of interest, thereby investigating the effect of the exposure on a specific outcome.^7,8 Due to the random assortment of single nucleotide polymorphisms (SNPs), they are less likely to be influenced by confounding factors. The MR design can mitigate residual confounding and reverse causation, strengthening the causal inference of the exposure’s association with the outcome. In this study, we conducted a two-sample MR analysis to assess the causal relationship between H. pylori infection and rosacea.

Materials and Methods

Data Sources

We retrieved GWAS data for H. pylori infection from MRC-IEU Open GWAS (https://gwas.mrcieu.ac.uk) and GWAS catalog (https://www.ebi.ac.uk/gwas/), and the GWASID were “ieu-b-4905” and “ebi-a-GCST90006910”, and we call them H. pylori infection (ieu) and H. pylori infection (ebi) separately in this study. Data for rosacea were retrieved from FinnGen database (https://r12.finngen.fi/), and the GWASID were “finngen_R12_L12_ROSACEANAS”. The study population consisted exclusively of individuals with European ancestry, potentially limiting the generalizability of our findings.

Instrumental Variants (IVs) Selection

We use Linkage-disequilibrium pruning to remove the influence of highly correlated single nucleotide polymorphisms (SNPs). These SNPs were clumped based on the linkage disequilibrium (LD) structure of European ancestry (r2 < 0.001, 10000 kb). We selected SNPs with P < 5×10⁻⁸ and found that the H. pylori dataset contained fewer than 3 SNPs in the MR analysis. Therefore, we adjusted the SNPs screening threshold to P < 5×10⁻⁶. Additionally, the F-statistics was calculated using the formula F = [(n-k-1)/k] × [R²/(1-R²)], where an F-value > 10 indicated sufficient strength of the instrumental variables. For datasets missing the effect allele frequency (eaf), the formula F = β²/SE² was applied for calculation. Conceptual framework diagram was shown in Figure 1.

Finally, 12 SNPs associated with “H. pylori infection (ieu)” and 11 SNPs related to “H. pylori infection (ebi)” were retrieved from GWAS. After removing SNPs for being unavailable or being palindromic with intermediate allele frequencies (rs7281117), there were 12 SNPs left as IVs for “H. pylori infection (ieu)” and 10 for “H. pylori infection (ebi)”.

Statistical Analysis

Inverse-variance weighted (IVW) method was the main method we used in this study. To make the results more accurate, MR Egger, simple mode, weighted median, and weighted mode were also performed by using the “TwoSampleMR” package in R (version 4.2.3). We used MR-PRESSO test to eliminate outliers and the heterogeneity was also carried out by Cochran’s Q test. And the reliability of the results is also proved by the method of leave-one-out.

Results

We conducted an MR analysis between H. pylori infection and rosacea. In the H. pylori infection (ieu) group, we identified 12 SNPs. All instrumental variables showed sufficient strength (F > 10), comfortably exceeding the weak instrument threshold. The IVW results indicated no genetic association between H. pylori infection and rosacea (OR = 0.894, 95% CI 0.70–1.013, P = 0.078). In the H. pylori infection (ebi) group, we obtained 10 SNPs. Similarly, the IVW analysis showed no significant genetic association (OR = 1.021, 95% CI 0.932–1.119, P = 0.655). Results were similar in the analysis based on the weighted median method: H. pylori infection (ieu) (OR = 0.98, 95% CI 0.83–1.17, P = 0.849), H. pylori infection (ebi) (OR = 1.0, 95% CI 0.88–1.13, P = 0.965). Meanwhile, results based on the simple mode: H. pylori infection (ieu) (OR = 1.04, 95% CI 0.78–1.38, P = 0.802), H. pylori infection (ebi) (OR = 0.95, 95% CI 0.79–1.15, P = 0.600), and results based on the weighted mode: H. pylori infection (ieu) (OR = 1.03, 95% CI 0.82–1.29, P = 0.817), and H. pylori infection (ebi) (OR = 0.97, 95% CI 0.83–1.15, P = 0.772). They all contribute to enhancing the credibility of this causal inference. The main results of this study are shown in Figure 2. And two scatters are shown in Figures 3 and 4.

Figure 2 Forest plot of MR results.

Figure 3 Scatterplot of SNP potential effects of H. pylori infection ebi-a-GCST90006910 to rosacea.

Figure 4 Scatterplot of SNP potential effects of H. pylori infection ieu-b-4905 to rosacea.

Additionally, the results of heterogeneity and pleiotropy are presented in Table 1. As for heterogeneity, no statistically significant findings were observed in either the H. pylori infection (ieu) group (p = 0.4840) or the H. pylori infection (ebi) group (p = 0.5188). As for the MR-Egger pleiotropy test, no statistically significant results were detected in either the H. pylori infection (ieu) group (p = 0.3908) or the H. pylori infection (ebi) group (p = 0.4308), indicating the reliability of the findings. Furthermore, MR-PRESSO did not identify any outliers, confirming the stability and robustness of the results.

Table 1 Sensitivity Analyses of MR

Discussion

Rosacea is a common, sometimes hereditary, chronic inflammatory facial skin condition that primarily affects adults, mostly in women.^9,10 Although the underlying causes of rosacea are not fully understood, several studies have suggested there is an association with H. pylori infection.^11,12 Some studies have also observed that symptoms in rosacea patients can improve following the eradication of H. pylori.^3,6,13 It has been reported that H. pylori can stimulate the immune system to produce a large number of inflammatory mediators, resulting in the occurrence and aggravation of rosacea inflammation.^14,15 H. pylori increases the synthesis of oxygen metabolites,^15,16 such as superoxide and pro-inflammatory cytokines, which can affect skin physiology, including vasodilation, inflammation and immune regulation.^14,17 And TLR2 and TLR4 can activate Toll-like receptors, leading to NF-κB pathway activation, along with nonspecific release of chemokines such as IL-8. Additionally, H. pylori infection may alter neuropeptide levels, impairing the Nrf2 antioxidant pathway and amplifying local inflammatory responses.^18,19 These mechanisms could potentially exacerbate rosacea. However, these shared pathways only indicate correlation and cannot establish a causal relationship.

In our study, we found that all methods used in this study suggested no association between H. pylori infection and rosacea, align with a previously published meta-analysis²⁰ concluded that there was weak correlation association between rosacea and H. pylori infection and that no significant improvement in rosacea skin symptoms was observed after H. pylori treatment. It seems like H. pylori is an aggravating factor rather than a cause. Moreover, we postulated that the lifestyle of H. pylori-infected individuals might potentially exert an influence on the outcomes of previous studies. We employed MR methodology, which effectively mitigated the influence of confounding factors through the utilization of genetic variants as instrumental variables.

We identified multiple H. pylori-related datasets in the GWAS database, among which only the Anti-H. pylori IgG dataset contained a sufficient number of SNPs, thus justifying our selection of these two specific datasets for this study. Additionally, our analysis was confined to northern European populations to ensure genetic homogeneity, as genetic variations across different ethnic groups could potentially confound the results. While this approach enhances the reliability of our findings within this specific population, it may introduce a degree of bias, limiting the generalizability of our conclusions. Future research should aim to include diverse racial and ethnic groups to explore the association between rosacea and H. pylori infection. Rosacea comprises multiple clinical subtypes.²¹ However, since current rosacea GWAS databases lack subtype classification and do not provide specific case numbers or proportions for each subtype, we were unable to conduct MR analyses between H. pylori infection and distinct rosacea subtypes—a significant limitation of this study. Nevertheless, our null findings do not preclude potential associations in specific subtypes, as substantial literature reports elevated H. pylori infection rates in papulopustular rosacea patients. Current studies on the association between H. pylori infection and rosacea have covered a wide range of regions, including Egyptian, Chinese, Russian, and German patients. Studies suggesting no significant association between H. pylori and rosacea^5,22 may also be influenced by factors such as: sample size limitations, heterogeneous composition of rosacea subtypes, geographical variations and sensitivity of detection methods. Future research should prioritize the following directions to address current limitations: 1. Precision subtype stratification of rosacea; 2. broader geographical representation; 3. standardized high-sensitivity detection protocols; 4. expand the sample size.

Conclusion

Our MR analysis provides no evidence of a causal relationship between H. pylori infection and rosacea. This indicates that patients with rosacea may not need routine testing for H. pylori infection and routine eradication of H. pylori may not benefit rosacea patients.

Ethics Approval and Consent to Participate

According to item 1 and 2 of Article 32 of the Measures for Ethical Review of Life Science and Medical Research Involving Human Subjects adopted by the National Science and Technology Ethics Committee of the People’s Republic of China dated February 18, 2023, since the data used in this study will not cause any harm to humans, does not involve any sensitive personal information or commercial interests, and the selected database is open and legal, approval from the ethics review committee is not required.

Funding

There is no funding for this study.

Disclosure

The authors declare that there is no conflict of interest.

References

1. Backert S, Neddermann M, Maubach G, Naumann M. Pathogenesis of Helicobacter pylori infection. Helicobacter. 2016;21:19–25. doi:10.1111/hel.12335

2. Sjomina O, Pavlova J, Niv Y, et al. Epidemiology of Helicobacter pylori infection. Helicobacter. 2018;23(1):e12514. doi:10.1111/hel.12514

3. Lazaridou E, Korfitis C, Kemanetzi C, et al. rosacea and Helicobacter pylori: links and risks. Clin Cosmet Invest Dermatol. 2018;2018:305–310. doi:10.2147/ccid.s121117

4. Xingzhe Y. Relationship between Helicobacter pylori and Rosacea: review and discussion. BMC Infect Dis. 2018;18(1):318. doi:10.1186/s12879-018-3232-4

5. Egeberg A, Weinstock LB, Thyssen EP, et al. Rosacea and gastrointestinal disorders – a population‐based cohort study. Br J Dermatol. 2016;176(1):100–106. doi:10.1111/bjd.14930

6. Agnoletti AF, Col ED, Parodi A, et al. Etiopathogenesis of rosacea: a prospective study with a three-year follow-up. G Ital Dermatol Venereol. 2016;152(5):418–423. doi:10.23736/S0392-0488.16.05315-3

7. Davey Smith G, Hemani G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet. 2014;23(R1):89–98. doi:10.1093/hmg/ddu328

8. Sanderson E, Smith GD, Windmeijer F, Bowden J. Corrigendum to: an examination of multivariable Mendelian randomization in the single-sample and two-sample summary data settings. Int J Epidemiol. 2020;49(3):1057. doi:10.1093/ije/dyaa101

9. Two AM, Wu W, Gallo RL, Hata TR. Rosacea: part I. Introduction, categorization, histology, pathogenesis, and risk factors. J Am Acad Dermatol. 2015;72(5):749–758. doi:10.1016/j.jaad.2014.08.028

10. Ahn CS, Huang WW. Rosacea Pathogenesis. Dermatol Clin. 2018;36(2):81–86. doi:10.1016/j.det.2017.11.001

11. Gravina A, Federico A, Ruocco E, et al. Helicobacter pylori infection but not small intestinal bacterial overgrowth may play a pathogenic role in rosacea. United Eur Gastroenterol J. 2015;3(1):17–24. doi:10.1177/2050640614559262

12. Albalbeesi A, Alsalman H, Alotaibi H, et al. Prevalence of Helicobacter pylori infection among rosacea and chronic spontaneous urticaria patients in a tertiary hospital in Riyadh, Saudi Arabia. Cureus. 2021. doi:10.7759/cureus.17617

13. Hejazi S, Aghaei M, Aghaei S, et al. Association between the treatment of rosacea and eradication of Helicobacter pylori infection. Adv Biomed Res. 2023;12(1):173. doi:10.4103/abr.abr_236_22

14. Alipour M. Molecular mechanism of Helicobacter pylori induced gastric cancer. Gastrointest Cancer. 2021;52(1):23–30. doi:10.1007/s12029-020-00518-5

15. Santos MLC, de Brito BB, da Silva FAF, et al. Helicobacter pylori infection: beyond gastric manifestations. World J Gastroenterol. 2020;26(28):4076–4093. doi:10.3748/wjg.v26.i28.4076

16. Guarneri C, Ceccarelli M, Rinaldi L, Cacopardo B, Nunnari G, Guarneri F. Helicobacter pylori and skin disorders: a comprehensive review of the available literature. Eur Rev Med Pharmacol Sci. 2020;24(23):12267–12287. doi:10.26355/eurrev_202012_24019

17. Colgecen E, Karacavus S, Kader C, Ozyurt K, Celikbilek M. Helicobacter pylori infection in multiple forms of dermatosis. Acta Med Mediterr. 2015;31(6):1127–1132.

18. Emiroglu N, Cengiz FP, Gulsel Bahali A, Biyik Ozkaya D, Kucuk OS, Onsun N. Role of Helicobacter pylori in Rosacea. Bezmialem Sci. 2018;6(1):6–8. doi:10.14235/bs.2018.1359

19. Geng RSQ, Bourkas AN, Mufti A, Sibbald RG. Rosacea: pathogenesis and therapeutic correlates. J Cutan Med Surg. 2025;28:8–12. doi:10.1177/12034754241229365

20. Jørgensen A-HR, Egeberg A, Gideonsson R, Weinstock LB, Thyssen EP, Thyssen JP. Rosacea is associated with Helicobacter pylori: a systematic review and meta-analysis. J Eur Acad Dermatol Venereol. 2017;31(12):2010–2015. doi:10.1111/jdv.14352

21. van Zuuren EJ, Arents BWM, van der Linden MMD, et al. Rosacea: new concepts in classification and treatment. Am J Clin Dermatol. 2021;22(4):457–465. doi:10.1007/s40257-021-00595-7

22. Bhattarai S, Agrawal A, Rijal A, Majhi S, Pradhan B, Dhakal SS. The study of prevalence of Helicobacter pylori in patients with acne rosacea. Kathmandu Univ Med J. 2012;10(40):49–52. doi:10.3126/kumj.v10i4.10995