Causation between the gut microbiota and inguinal hernia: a two-sample double-sided Mendelian randomization study

Abstract

Inguinal hernias are the most common type of enterocele and are frequently caused by defects in the abdominal wall muscles in the groin area. Numerous animal models and human studies have shown that the gut microbiota is associated with skeletal muscle aging and loss. However, the causation between the gut microbiota and inguinal hernia remains unclear. To reveal the causal association between the gut microbiota and inguinal hernia, we conducted a two-sample double-sided Mendelian randomization analysis. We used genome-wide association analysis (GWAS) summary statistics of the gut microbiota from the MiBioGen consortium and GWAS statistics of inguinal hernia from the FinnGen R10 database. The causation between the gut microbiota and inguinal hernia was explored through the inverse variance weighted (IVW) method, MR Egger regression method, weighted median method, weighted model method, and simple model method. Sensitivity analysis was used to test whether the Mendelian randomization analysis results were reliable. Reverse Mendelian randomization was used to conduct effect analysis and sensitivity analysis using the entire gut microbiota as the outcome. The IVW results indicated that Verrucomicrobia, Lactobacilliales, Clostridiaceae1, Butyricococcus, Categorybacter, Hungatella, Odoribacter, and Olsenella had a direct negative causation with the gut microbiota. The reverse Mendelian Randomization results showed that Eubacterium brachygroup, Eubacterium eligensgroup, Eubacterium xylanophilumgroup, Coprococcus3, Ruminococcus1, and Senegalimassilia were directly related to inguinal hernia. The bilateral sensitivity analysis revealed no heterogeneity or horizontal pleiotropy. The results confirmed that 8 bacterial traits had a negative causation with inguinal hernia. Reverse MR analysis revealed a positive correlation between inguinal hernia and 6 bacterial traits. Modulating the diversity and components of the gut microbiota is envisaged to contribute to improving the incidence and prognosis of inguinal hernia.

Introduction

An enterocele is a bulging of the peritoneum, peritoneal fat, or abdominal organs through congenital or acquired orifices in the abdominal wall^1–3. The probability of experiencing an inguinal hernia during one's lifetime is approximately 25% for males and 3% for females, and this probability increases in tandem with advancing years^4–6. The groin hernia encompasses three distinct categories distinguished by the location of the hernia sac neck concerning the inguinal (Hesselbach) triangle: direct inguinal, indirect inguinal, and femoral³. Direct inguinal hernia mostly occurs in elderly men. The anatomical substance protrudes through the posterior boundary of the inguinal canal, and the neck of the hernia sac is found medially to the inferior epigastric artery. Indirect inguinal hernia commonly occurs in children and young adults. The small intestine, omentum, and other tissues protrude through the internal ring of the inguinal canal, and the hernia sac neck is located outside the inferior epigastric artery⁷. Femoral hernia mainly occurs in women over 40 years old. The tissue protrudes along the femoral canal under the inguinal ligament. The hernia sac neck is located inside the femoral blood vessel. The treatment modalities for inguinal hernias often include hernia repair as a fundamental approach; 1.6 million inguinal hernias are diagnosed in the United States each year, and more than 500,000 surgeries are performed⁸.

The human intestinal environment harbors an immense population of bacteria, totaling trillions, and these microorganisms have undergone a coevolutionary journey alongside the human genome. The gut microbiota is involved in regulating the host body and is related to a variety of diseases⁹. Nevertheless, the constitution of the gut microbiota undergoes varies depending on the host species, and an array of internal and external factors, along with both biotic and abiotic influences, can induce changes in the intricate balance of the gut microbiota composition^10–15. Numerous research endeavors have delved into potential mechanisms through which the gut microbiota may contribute to the depletion of skeletal muscle mass, encompassing aspects such as protein anabolism, mitochondrial dysfunction, persistent inflammation, immune responses, and imbalances in metabolic processes¹⁶. In a groundbreaking study conducted in 2004, Backhed et al. injected cecal contents from conventionally raised animals into the intestines of germ-free mice¹⁷. The results indicated that the body fat of mice in the experimental group increased by 60%, and insulin sensitivity and glucose tolerance decreased accordingly. Since skeletal muscle is one of the tissues that processes glucose, this finding suggests microbial involvement in mediating the functional regulation of muscle metabolism. AMP-activated protein kinase (AMPK) and carnitine palmitoyltransferase-1 (CPT-1) activities were significantly greater in the skeletal muscle of germ-free mice than in that of mice harboring gut microbiota, suggesting greater oxidation ability. The findings of this investigation underscore that the gut microbiota possesses the capacity to shape the biological makeup by regulating bioenergetic pathways within skeletal muscle. Alterations in the constitution of the gut microbiota have the potential to contribute to the aging of muscles and the onset of sarcopenia^18–20. Within the framework of the musculoskeletal system, the gut microbiota assumes a pivotal role by intricately managing intestinal permeability, energy metabolism, hormone secretion, systemic inflammatory pathways, and immune responses²¹.

Mendelian randomization (MR) is a contemporary method for determining the causal relationships between the gut microbiota and the presence of inguinal hernia. MR relies on genetic variants as instrumental variables (IVs) to evaluate the causal relationship between exposure and the disease outcomes. Because genetic variation is assigned randomly and lacks any connection to offspring, the presence of genetic variation, along with subsequent outcomes, remains unaltered by potential confounding factors. This finding substantiates the authenticity of the established causal sequence^22,23. MR has been broadly employed to probe the causal connections between the gut microbiota and an array of health conditions, including metabolic diseases, irritable bowel syndrome, preeclampsia, and several other diseases^24–28.

Therefore, in this study, the gut microbiota and inguinal hernia were selected as the research objects, and two-sample double-sided MR analysis was used to evaluate the causation between the gut microbiota and inguinal hernia and provide new biological diagnostic markers, treatment strategies, and theoretical foundations for further research on the mechanism of inguinal hernia.

Methods

Design

The experimental design used a two-sample double-sided MR analysis to evaluate the causation between the gut microbiota and inguinal hernia. Initially, comprehensive GWAS summary statistics about the gut microbiota and inguinal hernia were acquired. Therefore, to ensure the credibility of experimental findings, it is imperative to satisfy the three fundamental assumptions of MR. Single-nucleotide polymorphisms (SNPs) of IVs are strongly associated with exposure factors (gut microbiota or inguinal hernia); SNPs are independent of identified confounding factors; SNPs affect outcome factors (inguinal hernia or gut microbiota) only through exposure factors (gut microbiota or inguinal hernia) (Fig. 1).

Fig. 1.

Schematic diagram of the core assumptions of Mendelian randomization.

Material

The GWAS summary statistics on the gut microbiota were obtained from the MiBioGen consortium and are accessible at https://mibiogen.gcc.rug.nl/. This consortium stands as the most extensive, multiethnic collaboration, conducting a genome-wide meta-analysis specifically focused on gut microbiota^29,30. The study included 18,965 individuals from 18 cohorts, most of whom had European ancestry (n = 13,266), targeting variable regions V4, V3–V4, and V1–V2 of the 16S rRNA gene to analyze the microbial composition and classify using direct taxonomy. Microbiota quantitative trait locus (mbQTL) mapping analysis was conducted to pinpoint host genetic variability correlated with genetic loci linked to the abundance levels of bacterial taxa within the gut microbiota. During the compilation of this article, 15 bacterial species lacking distinct nomenclature, consisting of 3 families and 12 genera whose identities remain unidentified, were deliberately excluded. Subsequently, a refined selection of 196 bacterial species (comprising 119 genera, 32 families, 20 orders, 16 classes, and 9 phyla) was meticulously chosen for MR analysis. The GWAS summary statistics for inguinal hernia were obtained from the FinnGen R10 database (https://www.finngen.fi/en), which included 35,248 inguinal hernia cases and 352,418 controls, all of which were from European populations³¹. Notably, the data for this study came from two different databases. The FinnGen R10 database meticulously examined the genetic ancestry data of participants from the study through rigorous genotyping quality control (QC), effectively eliminating outliers. This meticulous process ensured the absence of sample overlap between the individual datasets of the 18 cohorts comprising the MiBioGen consortium and the GWAS data sourced from the FinnGen R10 database.

The summary statistics employed in this publication originate from openly accessible databases that can be obtained without cost. All GWAS summary statistics included in the article were approved by the respective ethics agency.

Instrument variable selection

This study selected SNPs that were closely related to bacterial taxa (P < 1.0 × 10^–5) as IVs to obtain comprehensive data. Linkage disequilibrium (LD) analysis was performed based on European population sample data, with parameters set to r² < 0.001 and kb = 10,000. The robustness of IVs is assessed by computing the F statistic. An F value exceeding 10 signifies the absence of a weak IV error³². The IVs with F < 10 are eliminated. Finally, to prevent alleles from affecting the results, palindromic SNPs were removed through palindromic sequence detection.

r² is the variance of exposure explained by SNPs, and its calculation formula is as follows: $r^2=2\times {\beta }^2\times EAF\times \left(1,-,E,A,F\right)/2\times {\beta }^2\times EAF\times \left(1,-,E,A,F\right)+{\mathit{SE}}^2\times 2\times N\times EAF\left(1,-,E,A,F\right)$. Within the equation, EAF represents the frequency of the effect allele, β denotes the value of the allele's effect, and SE represents the standard error. The F statistic calculation formula is as follows: $F=\left[\left(N,-,k,-,1\right),/,k\right]\times r^2/\left(1,-,r^2\right)$. N is the number of samples in the exposure statistic, and k is the number of SNPs.

Sensitivity analysis

The robustness of the causation of the gut microbiota to inguinal hernia was measured through a series of sensitivity analyses. Cochran Q analysis can compute distinctions among IVs, and a P value less than 0.05 indicates the presence of heterogeneity. Depending on whether there was heterogeneity, the random effects model or fixed effects model was selected for analysis. Horizontal pleiotropy testing through MR Egger regression and MR-PRESSO analysis. The stability of the statistic was tested through leave-one-out analysis, which removes individual SNPs to determine whether there are SNPs that may have a strong effect.

The principal MR analysis in this research adopted the IVW method, while supplementary analyses, including MR Egger and the weighted median, were employed to enhance the depth of causal inference. If there was no horizontal pleiotropy in the data, there was no bias in the IVW results. Furthermore, this article employed reverse MR analysis to deduce the potential existence of reverse causation between the bacterial taxa identified in forward MR analysis and the occurrence of inguinal hernia.

The packages in R^33,34 and Stata³⁵ are facilitating the adoption and implementation of MR for two-sample summary data. The statistical method used in this study is causal analysis according to the latest and recognized MR analysis process. All statistical analysis procedures in this article were performed in RStudio software (version 4.3.2), using the TwoSampleMR and MR-PRESSO software packages for analysis.

Results

Causal effects of the gut microbiota on inguinal hernia

Following the screening criteria for IVs with a significance threshold of P < 1.0 × 10⁻⁵, a cumulative of 2,616 SNPs were acquired from a pool of 196 intestinal flora. These included 124 SNPs across 9 phyla, 223 SNPs spanning 16 classes, 279 SNPs in 20 orders, 444 SNPs within 32 families, and 1,546 SNPs associated with 119 genera. The F values corresponding to all SNPs were greater than 10. Therefore, this study is not susceptible to weak IV bias. MR analysis was performed on 196 bacterial taxa by the IVW method, and SNPs with P < 0.05 were screened out. The data analysis revealed that a total of 72 SNPs in 8 bacterial taxa are causally related to inguinal hernia (Fig. 2). These bacterial taxa included 1 phylum (12 SNPs), 1 order (15 SNPs), 1 family (10 SNPs), and 5 genera (35 SNPs), namely, Verrucomicrobia, Lactobacillales, Clostridiaceae1, Butyricococcus, Catenibacter, Hungathella, Odoribacter, and Olsenella.

Fig. 2.

Forest plot to evaluate the causal effect between gut microbiota and inguinal hernia using values obtained by the IVW MR method.

The results of the IVW analysis revealed that Verrucomicrobia (OR = 0.9029, 95% CI: 0.8375–0.9734, P = 0.0077), Lactobacillales (OR = 0.9087, 95% CI: 0.8378–0.9857, P = 0.0211), Clostridiaceae1 (OR = 0.9017, 95% CI: 0.8236–0.9871, P = 0.0251), Butyricococcus (OR = 0.8678, 95% CI: 0.7849–0.9594, P = 0.0056), Catenibacter (OR = 0.9211, 95% CI: 0.8562–0.9910, P = 0.0276), Hungathella (OR = 0.8748, 95% CI: 0.8133–0.9410, P = 0.0003), Odoribacter (OR = 0.8583, 95% CI: 0.7690–0.9580, P = 0.0064), Olsenella (OR = 0.9314, 95% CI: 0.8882–0.9767, P = 0.0034). MR study revealed a negative correlation between the abovementioned gut microbiota and inguinal hernia (Fig. 3).

Fig. 3.

Scatter plot of the causal relationship between gut microbiota and inguinal hernia.

The Cochran Q analysis results revealed no obvious heterogeneity among the selected IVs. The MR Egger analysis results suggested a lack of horizontal pleiotropy within the bacterial taxa examined in this study (Table 1). The outcomes of the leave-one-out method analysis indicate the absence of evident outliers among the chosen IVs, and the MR analysis results are reliable (Fig. 4).

Table 1.

Sensitivity analysis of the gut microbiota and inguinal hernia status.

Category	Exposure factors	Heterogeneity test		Horizontal pleiotropy test
		IVW Q P value	MR Egger Q P value	MR Egger intercept value	MR Egger intercept P value
Phylum	Verrucomicrobia	0.496	0.419	0.003	0.728
Order	Lactobacillales	0.735	0.810	– 0.011	0.195
Family	Clostridiaceae1	0.971	0.964	0.006	0.568
Genus	Butyricicoccus	0.608	0.503	– 0.003	0.755
Genus	Catenibacterium	0.459	0.403	– 0.053	0.471
Genus	Hungatella	0.693	0.526	– 5.62E-05	0.999
Genus	Odoribacter	0.450	0.671	– 0.021	0.169
Genus	Olsenella	0.344	0.462	– 0.016	0.152

Fig. 4.

Leave one out analysis of gut microbiota and inguinal hernia.

Reverse MR analysis

In the reverse MR analysis involving 196 intestinal flora and inguinal hernias, the results point toward a forward causation link between 6 intestinal flora and the manifestation of inguinal hernia (Fig. 5). A direct causative link is established between inguinal hernia and Eubacterium brachygroup, Eubacterium eligensgroup, Eubacterium xylanophilumgroup, Coprococcus3, Ruminococcus1, and Senegalimassilia (Fig. 6). The outcomes of sensitivity analysis in the reverse MR analysis revealed the absence of heterogeneity and horizontal pleiotropy.

Fig. 5.

Forest plot to evaluate the reverse causal effect between gut microbiota and inguinal hernia using values obtained by the IVW MR method.

Fig. 6.

Scatter plot of reverse causality between gut microbiota and inguinal hernia.

Discussion

This study employs a two-sample double-sided MR analysis to assess the causal association between gut microbiota and the occurrence of inguinal hernia. Our comprehensive analysis unveiled intriguing insights into the protective effects exerted by specific microbial entities against the onset of inguinal hernia. Notably, Verrucomicrobia, Lactobacillales, Clostridiaceae1, Butyricococcus, Catenibacter, Hungathella, Odoribacter, and Olsenella were found to exhibit a discernible protective role in mitigating the development of inguinal hernia. This observation underscores the potential significance of these microbial taxa in conferring resilience against inguinal hernia pathology.

Numerous observational studies have documented an association between gut microbiota composition and the incidence of inguinal hernia. Our findings are consistent with previous studies indicating that Lactobacilliales, Butyricococcus and Categorybacter are correlated with a reduced risk of inguinal hernia recurrence^36–38. Lactobacilliales, recognized as a probiotic, has been demonstrated to be associated with diverse health advantages, including antibacterial properties, modulation of cholesterol metabolism, immune modulation, antioxidant effects, antidiabetic effects, alleviation of allergic responses, and suppression of tumorigenesis³⁹. Through the utilization of multiview learning approaches on fecal metagenomic data, Chen et al. established a human gut microbiota aging clock, implying that the composition of the gut microbiota changes with advancing age⁴⁰. The insulin/insulin-like growth factor-1 (IGF-1) signaling cascade triggers the activation of the mammalian target of rapamycin (mTOR), leading to an increase in protein synthesis for the preservation of muscle mass⁴¹. Reduced insulin/IGF-1 signaling in older adults contributes to an increase in insulin resistance, ultimately initiating the depletion of skeletal muscle mass⁴². Cani et al. demonstrated that feeding mice a high-fat diet impaired the tight junctions of the intestinal epithelium and elevated intestinal permeability, indicating that a high-fat diet can induce leakage of lipopolysaccharide (LPS), an external membrane element found in gram-negative bacteria, from the gut into the circulation⁴³. In individuals inoculated with LPS, there was a substantial increase in the binding activity of nuclear factor-kappa B (NF-κB) and the phosphorylation of c-Jun N-terminal kinase (JNK) within skeletal muscle, demonstrating that endotoxin is involved in mediating the process of impaired skeletal muscle glucose tolerance.

A cohort study of 266 patients found that elevated serum triglyceride levels were an independent predictor of inguinal hernia⁴⁴. Wang, Y. et al. elucidated that Lactobacilliales administration significantly reduces serum levels of total cholesterol, low-density lipoprotein cholesterol, and triglycerides in high-fat-fed mice⁴⁵. Dysfunction in intestinal barrier integrity can elevate the release of LPS by the gut microbiota into the bloodstream⁴⁶, consequently impairing skeletal muscle glucose tolerance and predisposing to skeletal muscle aging and atrophy. Lactobacilliales exhibits the capability to sustain intestinal barrier integrity by modulating blood lipid profiles and averting the translocation of LPS into systemic circulation. Certain probiotic strains have been documented to induce mucin-3 expression in intestinal epithelial cells⁴⁷, promoting mucus production⁴⁸ and thereby contributing to the preservation of intestinal barrier integrity. Slattery, C. et al. observed that specific strains of Lactobacilliales can reduce cellular activation of NF-κB by approximately fifty percent³⁹. These previous studies indicate that Lactobacilliales can also interfere with endotoxin-mediated impairment of glucose tolerance in skeletal muscle by inhibiting NF-κB activation. In addition, some studies have reported that as age increases, the microecology of the gut microbiota changes, and the decrease in the proportion of Verrucomicrobia can lead to glucose metabolism dysfunction in the human body⁴⁹. This is consistent with our research. A normal proportion of probiotics such as Verrucomicrobia will maintain normal skeletal muscle glucose metabolism and delay the aging and atrophy of skeletal muscle.

Reduced abdominal wall strength and increased intra-abdominal pressure are the two main causes of inguinal hernia⁵⁰. This study further revealed the gut microbiota and metabolites may affect inguinal hernia through multiple mechanisms. Intestinal commensal flora has been implicated in the modulation of various physiological processes. It has been shown to lower blood lipid and blood sugar levels, promote the expression of intestinal epithelial mucin 3, and facilitate mucus regeneration. These actions contribute to the preservation of intestinal barrier integrity and attenuate LPS translocation into the bloodstream. Furthermore, intestinal commensal flora exhibits inhibitory effects on NF-κB activation, thus directly impeding the progression of impaired skeletal muscle glucose tolerance mediated by LPS. These multifaceted mechanisms collectively delay the onset of skeletal muscle aging and atrophy, consequently diminishing the risk of inguinal hernia. However, additional randomized controlled trials are warranted to validate and corroborate these observations.

With the development of GWAS, the data volume and measurement accuracy of genetic variation have continued to expand, significantly reducing data bias in the research process and laying the foundation for research on the causation between the gut microbiota and inguinal hernia. Because genetic variation is determined before birth, measurable genetic variables are not affected by environmental factors, making causal effect studies more reliable than observational studies and equivalent to natural randomized controlled trials. Therefore, with this research method, it is more reliable to explore the causation between the gut microbiota and inguinal hernia.

Our investigation employed information extracted from the most extensive GWAS meta-analysis of gut microbiota conducted by the MiBioGen consortium. Additionally, GWAS data on inguinal hernia, made available by the FinnGen R10 database, were utilized to perform a two-sample double-sided MR analysis. Using the gut microbiota as the exposure and inguinal hernia as the outcome, Verrucomicrobia, Lactobacilliales, Clostridiaceae1, Butyricococcus, Categorybacter, Hungatella, Odoribacter, and Olsenella were found to be directly negatively related to the gut microbiota. Regarding inguinal hernia as the primary exposure and gut microbiota as the corresponding outcome, Eubacterium brachygroup, Eubacterium eligensgroup, Eubacterium xylanophilumgroup, Coprococcus3, Ruminococcus1, and Senegalimassilia were directly related to inguinal hernia.

Currently, there is a lack of reported studies examining the relationship between the gut microbiota and inguinal hernia. This article is the first to use MR to analyze the causation between them. Compared with individual-level experimental studies, there are differences between traditional laboratory animals and the natural environment, which cannot reflect the real living conditions of humans. Therefore, there are limitations in studying the relationship between the gut microbiota and inguinal hernia from a clinical cross-sectional perspective. Using GWAS data from the large-scale MiBioGen consortium and the FinnGen R10 database to probe genetic data on the gut microbiota and inguinal hernia can improve the statistical power of causal associations. MR studies can overcome the effects of potential confounding and inverse causation, avoid waste of resources, and evaluate the potential causative relationship between the gut microbiota and inguinal hernia considering the genetic dimension of the host. Nevertheless, it is crucial to recognize certain limitations that exist within the framework of this study. The GWAS data of inguinal hernia used in this study cannot explore potential nonlinear relationships or stratification effects caused by differences in age, sex, health status, etc., which may cause heterogeneity. Due to issues with GWAS data on the gut microbiota and inguinal hernia, this study set the threshold at P < 1 × 10⁻⁵, and there may be some unavoidable confounding factors. Since GWAS data in public databases only include people of European ancestry, taking into account population stratification issues, this conclusion may not apply to non-European populations, and more future studies on the gut microbiota and inguinal hernia are needed for further verification.

Conclusions

This article presents the first extensive examination of the potential causation between the gut microbiota and inguinal hernia. In conclusion, this MR study involving two distinct datasets identified a causal relationship between Lactobacilliales and Verrucomicrobia and inguinal hernia. However, randomized controlled trials are essential to elucidate the protective mechanisms of intestinal commensal bacteria against inguinal hernia and to establish their specific protective effects. Reverse MR analysis demonstrated a direct and positive correlation between six intestinal microbiota phenotypes and inguinal hernia. Nevertheless, further investigation is imperative to elucidate the reciprocal influence of inguinal hernia on intestinal microecology. Based on this article and previous studies, the biological effects of intestinal commensal bacteria in delaying skeletal muscle aging and atrophy by maintaining normal blood sugar and lipid levels, as well as its positive clinical significance in reducing the risk of inguinal hernia, may serve as potential focal points and research directions for future clinical studies and randomized controlled trials.

Abbreviations

GWAS

Genome-wide association analysis

IVW

Inverse variance weighted

MR

Mendelian randomization

AMPK

AMP-activated protein kinase

CPT-1

Carnitine palmitoyltransferase-1

IVs

Instrumental variables

SNPs

Single-nucleotide polymorphisms

mbQTL

Microbiota quantitative trait locus

QC

Quality control

LD

Linkage disequilibrium

IGF-1

Insulin/insulin-like growth factor-1

mTOR

Mammalian target of rapamycin

LPS

Lipopolysaccharide

NF-κB

Nuclear factor-kappa B

JNK

C-Jun N-terminal kinase

Author contributions

H provided guidance on data analysis methods and manuscript writing. C and Y collected and analyzed the GWAS data for the gut microbiota and inguinal hernia, and were major contributors in writing the manuscript. All authors read and approved the final manuscript.

Data availability

The datasets generated during and analyzed during the current study are available in the MiBioGen consortium (https://mibiogen.gcc.rug.nl/) and FinnGen R10 database (https://www.finngen.fi/en).

Competing interests

The authors declare no competing interests.