Causal associations between Helicobacter pylori infection and pregnancy and neonatal outcomes: a two-sample Mendelian randomization study

Helicobacter pylori (H. pylori) is a gram-negative bacterium with urease, catalase, and oxidase activity that colonizes the human stomach (Zamani et al., 2018). It is one of the most common pathogens in the world, infecting more than half of the whole population (Kamboj et al., 2017). Therefore, the health impact of H. pylori infection, such as peptic ulcer disease, chronic gastritis, gastric adenocarcinoma, and gastric cancer, is crucial for public health (Kusters et al., 2006; McColl, 2010). Among pregnant women, the prevalence of H. pylori infection remains high in many countries (Baingana et al., 2014; Mubarak et al., 2014) and it has been suggested that increased susceptibility to H. pylori infection may be due to pregnancy itself (Lanciers et al., 1999).

A systematic review of studies published up to November 17th, 2018, exploring associations of H. pylori infection with pregnancy and neonatal complications, showed significantly increased risks of preeclampsia, gestational diabetes mellitus, spontaneous miscarriage, and low birthweight (Zhan et al., 2019). Consistently, some other studies also reported an adverse effect in certain outcomes (Wanyama et al., 2016; den Hollander et al., 2017; Li et al., 2020; Tang et al., 2021). In particular, for adverse pregnancy outcomes, H. pylori infection is associated with gestational diabetes mellitus (Li et al., 2020; Tang et al., 2021) and preeclampsia (Tang et al., 2021), while for adverse neonatal outcomes, H. pylori infection is associated with low birthweight (den Hollander et al., 2017; Wanyama et al., 2016) and small for gestational age (den Hollander et al., 2017). However, most single studies examined only one or few outcomes without a comprehensive evaluation. In addition, these observational studies were vulnerable to residual confounding, and variations in confounder control could cause heterogeneity between studies, thereby leading to controversial results. Thus, reexamining the impact of infections on a range of pregnancy and neonatal outcomes is essential to safeguard the health of both pregnant women and fetus.

Mendelian randomization (MR) is an epidemiological approach that reveals causality in an unbiased manner, relying on genetic variation as the instrumental variable (IV) to assess whether an exposure leads to a corresponding outcome (Lawlor et al., 2008). Since alleles segregate according to Mendel’s second law of inheritance and genotypes are randomly assigned from parent to offspring without the influence of confounding, the causal sequence is reasonable (Burgess and Thompson, 2021). Given the infeasible conduction of randomized controlled trials, a number of MR studies have assessed the correlation between gut microbiota and pregnancy outcomes (Li C. et al., 2022; Li P. et al., 2022), while H. pylori was underexplored as the most common gastrointestinal pathogen.

In this study, we performed a two-sample MR analysis to evaluate the causal associations between H. pylori infection and nine pregnancy and neonatal outcomes, namely, miscarriage, preeclampsia or eclampsia, gestational diabetes mellitus, placental abruption, premature rupture of membranes, postpartum haemorrhage, birthweight, gestational age, and preterm birth.

According to the selection criteria of IVs, we identified a total of 20 SNPs for anti-H. pylori IgG levels. All F-statistics were above 10. The detailed information is shown in Supplementary Table 2.

Under the IVW method, genetically predicted IgG levels were significantly associated with increased risks of preeclampsia or eclampsia (odds ratio [OR] = 1.12, 95% confidence interval [CI] 1.01–1.24, P = 0.026) and premature rupture of membranes (OR = 1.17, 95% CI 1.05–1.30, P = 0.004). For preeclampsia or eclampsia, similar results were obtained via MR-Egger (OR = 1.32, 95% CI 1.06–1.64, P = 0.027). For premature rupture of membranes, the OR estimates obtained from the weighted median (OR = 1.22, 95% CI 1.06–1.41, P = 0.006) were also consistent with those from IVW. No significant associations were observed in sporadic miscarriage, gestational diabetes mellitus, placental abruption, postpartum haemorrhage, birthweight, gestational age, and preterm birth. The results of the five MR analysis methods are displayed in Table 1.

Across the MR study, no evidence of directional pleiotropy and heterogeneity was found in the MR-PRESSO global test, MR-Egger intercept test, and Cochran’s IVW and MR-Egger Q tests, except for analyzing the causal relationship between H. pylori infection and birthweight (Table 2). The results of the leave-one-out permutation analysis showed that the overall risk estimate was not driven by certain SNPs (Figure 2). In addition, potential outliers and the effects of SNPs, individually and jointly, from each MR method were shown in scatter plots (Figure 3), which displayed a similar trend toward a positive MR association of anti-H. pylori IgG levels with preeclampsia or eclampsia and premature rupture of membranes.

In this MR study, we found that anti-H. pylori IgG level was causally associated with preeclampsia or eclampsia and premature rupture of membranes, but not associated with other pregnancy (miscarriage, gestational diabetes mellitus, placental abruption, and postpartum haemorrhage) and neonatal (birthweight, gestational age, and preterm birth) outcomes.

Emerging evidence has shown the role of H. pylori infection in preeclampsia-eclampsia. In 2006, Ponzetto et al. (Ponzetto et al., 2006) first reported that pregnant women with preeclampsia had a 19.2% higher rate of H. pylori seropositivity compared with uncomplicated women. Following that, other groups also demonstrated an epidemiological association between H. pylori infection and preeclampsia, especially for women who were infected with cytotoxin-associated gene A (CagA) positive strains (Mosbah and Nabiel, 2016; Bellos et al., 2018; Nourollahpour Shiadeh et al., 2019). In vitro studies further showed that anti-CagA antibodies could cross-react with cytotrophoblast cells through β-actin, thus reducing their invasiveness by decreasing ERK 1/2 activation, NF-kB translocation and MMP-2 expression (Franceschi et al., 2012). Since trophoblast invasion of maternal decidua is vital for embryo implantation and placental development, the infection-induced autoimmunity may lead to inadequate placentation and preeclampsia onset (Tersigni et al., 2014). In addition, Di Simone et al. (Di Simone et al., 2017) found that H. pylori infection was associated with abnormality of uterine arteries Doppler velocimetry in preeclamptic women, and anti-H. pylori IgG fractions from these women could inhibit endothelial cells’ proliferation, migration and differentiation both in vitro and in vivo. Therefore, H. pylori infection may also have an impact on the angiogenesis and vascular resistance, which constitutes an important mechanism of preeclampsia pathogenesis (Chaiworapongsa et al., 2014).

For the correlation with premature rupture of membranes, the mechanisms are still unclear but may be associated with systematic and local effects of H. pylori infection. On the one hand, H. pylori could stimulate the release of pro-inflammatory cytokines, such as interleukin (IL)-1β, IL-6, IL-8, tumor necrosis factor-α, and macrophage migration inhibitory factor (Xia et al., 2005; UstUn et al., 2010; Chen et al., 2022). Among infected patients, systemic indices of inflammation were also observed to be elevated, including white blood cell count and C-reactive protein (Graham et al., 1998; Oshima et al., 2005; UstUn et al., 2010). Given the crucial role of the inflammation-oxidative stress axis in fetal membrane weakening (Menon and Richardson, 2017), the infection may thus lead to premature rupture of membranes. On the other hand, research found that extracellular vesicles (EVs) could be derived from H. pylori-infected gastric epithelial cells and entered the blood circulation (Xia et al., 2020). Additionally, outer membrane vesicles (OMVs) released by H. pylori also existed in the serum samples (Park and Tsunoda, 2022). Both EVs and OMVs could serve as transport vehicles to deliver pathogenic virulence factors (e.g., CagA) to extragastric organs including brain (Qiang et al., 2022; Xie et al., 2023). In this regard, the fetal membrane may be directly affected as well, while further studies are warranted for investigation.

Consistent with previous pooled results (Tang et al., 2021), our study did not support the link between H. pylori infection and preterm birth. However, several cohorts have shown an increased risk of miscarriage (Hajishafiha et al., 2011), gestational diabetes mellitus (Cardaropoli et al., 2015; Li et al., 2020; Tang et al., 2021), and low birth weight (Wanyama et al., 2016; Grooten et al., 2017; Zhan et al., 2019) among infected pregnant women, which were not detected by the current MR analysis. This may be due to the insufficiency of cases as well as residual confounding of observational design. As for placental abruption and postpartum haemorrhage, no relevant research is available thus far and their associations with H. pylori infection remain to be explored.

To our knowledge, this is the first MR study on the association of H. pylori infection with a series of pregnancy and neonatal outcomes, thus eliminating the interference of confounding factors and reverse causation. To avoid sample overlap and its associated bias, we used the exposure and outcome datasets from multiple independent GWAS. An iterative MR analysis with five different approaches was conducted to acquire conservative results, and the absence of pleiotropy and heterogeneity in most sensitivity analyses further rule out the false-positive likelihood.

There are some limitations of the current study that should be acknowledged. Firstly, in order to minimize demographic bias, we selected only data from people of European descent, making the generalizability of our finding to other ethnic populations compromised. Secondly, the GWAS sample size of H. pylori infection was relatively small. Based on the traditional GWAS significance threshold (P <5×10⁻⁸), the SNPs obtained were too few for further study. Therefore, we used the locus-wide significance level (P <1×10^-5) for SNP selection, which may introduce weak instrument bias to the overall estimates. Thirdly, our analysis was based on summary statistics instead of raw data, and it was not possible to conduct further subgroup analyses on the specific strain of H. pylori (e.g., CagA-positive and CagA-negative) and severe degree or subtype of disease (e.g., early-onset and late-onset preeclampsia). Lastly, serological tests measuring the overall antibody immune response towards H. pylori are incapable of distinguishing ongoing from past infection. For a more specific differentiation, additional information is needed such as ¹³C-urea breath test, histological examination, or eradication record (Sabbagh et al., 2019). Therefore, the present study using genetically-predicted H. pylori seropositivity only represent the susceptibility to infection, and further studies are needed to compare the outcomes of different infection status.

In summary, this two-sample MR study showed that genetically predicted H. pylori infection was causally associated with increased risks of preeclampsia-eclampsia and premature rupture of membranes. Our findings confirm the epidemiological evidence on the adverse impact of H. pylori in pregnancy. Further studies are needed to thoroughly elucidate the pathophysiological mechanisms and assess the effectiveness of pre-pregnancy screening and preventive eradication.

All GWAS summary data used are publicly available. The download links are detailed in. 1

Ethical approval was not required because this study was based solely on publicly available GWAS summary-level statistics and no individual-level data were used.

JH: Formal analysis, Funding acquisition, Writing – original draft. YL: Formal analysis, Writing – original draft. DX: Investigation, Writing – review & editing. MC: Data curation, Writing – review & editing. QX: Data curation, Writing – review & editing. JC: Data curation, Writing – review & editing. LX: Data curation, Writing – review & editing. LY: Data curation, Writing – review & editing. QW: Data curation, Writing – review & editing. ZL: Conceptualization, Project administration, Writing – review & editing. JW: Conceptualization, Supervision, Writing – review & editing. LT: Conceptualization, Supervision, Writing – review & editing.

The authors express their gratitude to all participants and investigators of GWASs for their contributions and data sharing.

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This work was supported by the National Natural Science Foundation of China (82260315), Natural Science Foundation of Jiangxi Province (20224BAB216025), and Central Funds Guiding the Local Science and Technology Development (20221ZDG020071).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcimb.2024.1343499/full#supplementary-material↗