Causal Associations Between the Gut Microbiome and Aortic Aneurysm: A Mendelian Randomization Study

Abstract

Background: Observational studies have indicated an association between the gut microbiota and the occurrence and progression of aortic aneurysm (AA). However, the causal relationship between the gut microbiota and AA and its subtypes remains unclear. This study used Mendelian randomization (MR) to gain new insights into the relationship between the gut microbiota and AA, including AA subtypes.

Methods: We used summary data from a genome-wide association study of gut microbiota to determine genetically predicted microbial taxa. Additionally, we predicted causal relationships between the gut microbiota and AA, including AA subtypes. MR was conducted with two-sample MR with the inverse variance weighting, MR-Egger, weighted median, and weighted mode methods to assess the causal relationships. Heterogeneity and pleiotropy were evaluated with the MR-Egger method, Cochran’s Q test, and the MR-PRESSO Global test. The strength of the causal relationships between exposures and outcomes was assessed with Bonferroni correction. The stability of the MR results was evaluated with leave-one-out analyses. Reverse MR analysis was also performed to examine reverse causality.

Results: Through MR analysis, after Bonferroni correction, specific microbial taxa were found to have a causal relationship in AA and its subtypes. Specifically, the phylum Lentisphaerae (OR = 0.82, P = 0.001), class Lentisphaeria (OR = 0.81, P = 0.0028), and family Bifidobacteriaceae (OR = 0.79, P < 0.001) were negatively associated with AA risk, whereas the genus Family XIII UCG001 (OR = 1.33, P < 0.001) was positively associated with AA risk. Regarding subtypes, elevated levels of the genus Bilophila (OR = 1.36, P < 0.001) were closely associated with abdominal aortic aneurysm (AAA) occurrence. Lower levels of the family Bifidobacteriaceae (OR = 0.71, P < 0.001) and phylum Lentisphaerae (OR = 0.81, P = 0.0025), and higher levels of the genus Ruminococcaceae UCG014 (OR = 1.30, P < 0.001) exhibited strong causal relationships with thoracic aortic aneurysm (TAA).

Conclusion: Our study suggests that specific components of the gut microbiota have causal effects, either beneficial or detrimental, on AA risk, thus providing potentially valuable biomarkers for early diagnosis and potential therapeutic targets.

Main article text

Introduction

Aortic aneurysm (AA) is a localized, progressive degenerative processes of the aortic wall that leads to fixed dilatation to a diameter exceeding 1.5-fold the normal adjacent segment or vessel diameter; AA most commonly occurs in the infrarenal aorta [1]. Thoracic aortic aneurysm (TAA) and abdominal aortic aneurysm (AAA) are the most common types of AA encountered in clinical practice [2]. Although they have similar physical characteristics, their pathophysiological changes are not exactly identical, and AAA has a higher incidence rate [3, 4]. Previous epidemiological investigations have shown that the prevalence of AAA in older men varies, ranging from 6.2% in Italy to 1.7% in England [5]. In the United States, AAA alone has become one of the top 20 causes of death and a serious medical burden [6]. Although extensive research has examined the risk factors for AA, such as hypertension, advanced age, and sex, the gut microbiota is a new research direction.

The human gut microbiota is a complex system that may influence the progression of diseases, such as atherosclerosis, by regulating metabolic and immune modulation pathways [7]. Changes in gut microbial diversity have also been observed in populations with diabetes and hypertension [8, 9]. Recent studies have suggested that the gut microbiota may influence AA occurrence and progression. For example, gut microbiota metabolites can lead to aneurysm formation through mechanisms involving inflammation and apoptosis [10]. R. intestinalis and its metabolites affect the progression of AAA by influencing the formation of neutrophil extracellular traps [11]. Additionally, alterations in the gut microbiota have been found to affect the occurrence or progression of intracranial aneurysms [12, 13] or AAA [14]. These findings suggest that a specific gut microbiota may play major roles in the occurrence and development of AA and its subtypes. However, owing to various biases in traditional epidemiological studies and confusion due to reverse causality, the causal relationship between a particular gut microbiota and AA remains unclear.

Mendelian randomization (MR) is a statistical method based on whole-genome sequencing data used to uncover causal relationships. On the basis of the principles of MR, different genotypes determine different intermediate phenotypes. If these phenotypes represent specific exposure characteristics of individuals, then the association effects between genotypes and diseases can represent the exposure effect of the factor on the disease. Because allelic genes follow the principles of random allocation, this effect is not influenced by the confounding factors and reverse causal associations commonly found in traditional epidemiological studies. Therefore, MR provides a method to explore causal relationships between exposure and outcomes while minimizing the influence of confounding factors to the greatest extent [15]. The objective of this study was to investigate the causal relationship between the genetically predicted gut microbial composition and AA, including the subtypes TAA and AAA, through a two-sample MR approach. Our findings may provide guidance for the development of useful microbial biomarkers as potential therapeutic targets for AA and its subtypes.

Methods

Figure 1 illustrates the analytical flow of this study.

Figure 1

Study Design for MR Analysis of the Associations Between the Gut Microbiota and Aortic Aneurysm.

MR, Mendelian randomization; AA, aortic aneurysm; TAA, thoracic aortic aneurysm; AAA, abdominal aortic aneurysm.

Data Source for Gut Microbiota

The comprehensive genome-wide association study (GWAS) data regarding the composition of the human gut microbiome came from a large-scale multi-ethnic GWAS meta-analysis. The MiBioGen study, the largest and most diverse research aimed at studying host-genetics-microbiome associations, involves 18,473 individuals from 24 European, East Asian, American, and other cohorts. Multiple hypervariable regions (i.e., V1-V2, V3-V4, and V4) in the 16S ribosomal RNA (rRNA) gene were used to investigate the composition of the gut microbiota. A total of 122,110 single nucleotide polymorphisms (SNPs) in the 16S fecal microbiome were analyzed in 18,473 individuals [16, 17]. After exclusion of 15 microbial taxa without specific species names, a total of 196 bacterial groups of the 211 total bacterial taxa were included.

Data Source for Outcomes

The GWAS summary statistics for AA were obtained from FinnGen version 9 (https://www.finngen.fi/en), a medical project aimed at analyzing genomic and health data for participants in the Finnish biobank to identify genotype-phenotype correlations. The model included factors such as sex, age, and ten principal components as covariates [18]. The analysis process is detailed in the following link: (https://finngen.gitbook.io/documentation/methods/phewas). The study population includes 349,539 controls, 7395 patients with AA, 3548 patients with AAA, and 3510 patients with TAA. AA included ruptured thoracic aortic aneurysms; thoracic aortic aneurysms without mention of rupture; ruptured abdominal aortic aneurysms; abdominal aortic aneurysms without mention of rupture; ruptured thoracoabdominal aortic aneurysms; thoracoabdominal aortic aneurysms without mention of rupture; ruptured aortic aneurysms at unspecified sites; and aortic aneurysms at unspecified sites, without mention of rupture. AAA included ruptured abdominal aortic aneurysms and abdominal aortic aneurysms without mention of rupture. TAA included ruptured thoracoabdominal aortic aneurysms and thoracoabdominal aortic aneurysms without mention of rupture. The diagnosis of AA and its subtypes was made with nationwide registries harmonized over International Classification of Diseases (ICD) revisions 8, 9, and 10; cancer-specific ICD-O-3; and NOMESCO procedure codes (https://www.finngen.fi/en/researchers/clinical-endpoints).

Selection of Instrumental Variables

To ensure adequate selection of instrumental variables (IVs), we chose SNPs with P-values below the genome-wide significance threshold (1 × 10⁻⁵). Furthermore, we evaluated and excluded IVs with an F-statistic (formula: , where N represents the sample size) <10, to ensure the strength of the association between IVs and the exposure [19]. Additionally, to ensure the independence of SNPs, we set the linkage disequilibrium parameter (R²) of SNPs to 0.001 and considered a genetic distance of 10,000 kb. To prevent potential pleiotropy, we used PhenoScanner [20] to examine potential confounders associated with IVs, such as smoking and atherosclerosis, to mitigate their influence on the relationship between the exposure and outcomes (Supplementary Table 5). If the number of SNPs in the IVs selected by a certain microbial community was less than 2, the microbial community data were not included in the final results, because methods such as MR-Egger and leave-one-out could not be used for validation.

Statistical Analysis

Before conducting MR analysis, we first harmonized the IV SNPs. In this two-sample MR analysis, to investigate the causal relationship between the exposure factor and the outcome, we used the inverse variance weighted (IVW), MR-Egger, weighted median, and weighted mode methods. IVW is a classical MR analysis method that combines Wald ratio estimates of each IV into a meta-analysis. The MR-Egger method is based on the assumption of the Instrument Strength Independent of Direct Effect (InSIDE) and considers a certain degree of horizontal pleiotropy in the statistical process [21]. The weighted median allows for the inclusion of certain genetic variants that may be invalid [22]. The weighted mode enables certain IVs to not meet the requirements of MR inference when most IVs have similar causal estimates. Among these methods, the random-effects IVW method is the primary analysis method, which is used to perform a meta-analysis of SNP-specific Wald ratios, on the basis of an assumption of balanced horizontal pleiotropy [23]. If the above results are inconsistent, priority should be given to the IVW results as the primary result. The design of this study followed the STROBE-MR guidelines [24].

Sensitivity Analysis

Several statistical tests were used to examine the presence of violations of the assumptions of MR, particularly horizontal pleiotropy. First, Cochran’s Q statistic was calculated to quantify heterogeneity in the effects generated by IVs. Second, the MR-Egger intercept was assessed for deviation from zero to detect horizontal pleiotropy [19]. MR Pleiotropy Residual Sum and Outlier (MR-PRESSO) tests were applied to test for horizontal pleiotropy and outliers. Compared with MR-Egger, MR-PRESSO has greater accuracy in identifying horizontal pleiotropy and outliers [25]. In cases in which the MR-Egger and MR-PRESSO results were inconsistent, priority was given to the MR-PRESSO results as the main findings. Additionally, leave-one-out analysis was conducted to detect any heterogeneity driven by individual SNPs. The MR-Steiger test was used to evaluate and filter out SNPs that might indicate reverse causality relationships [26]. Furthermore, reverse causality analysis was performed to examine reverse causality relationships. To account for multiple testing, we applied Bonferroni correction with significance thresholds set at for each category level, for example, class . Statistical analysis was performed in R software version 4.2.3 (https://www.r-project.org/).

Results

Selection of Instrumental Variables for Gut Microbiomes

SNPs were selected as IVs within the gene locus range (P < 1 × 10⁻⁵). After removal of SNPs with low linkage disequilibrium effects and weak associations with the exposure factor (F < 10), a total of 2031 SNPs were retained and included in the analysis. These SNPs included 102 SNPs from 9 phyla, 179 SNPs from 16 classes, 217 SNPs from 20 orders, 341 SNPs from 31 families, and 1192 SNPs from 119 genus. The main information collected for these SNPs included the effect allele, other allele, standard error, P value, and other relevant data for analysis purposes.

Causal Effects of the Gut Microbiota on AA and its Subtypes

Two-sample MR analysis was used to identify causal relationships between the gut microbiota and aneurysms. Higher genetic predictions of the genus Family XIII UCG001 and genus Parasutterella were associated with greater risk of AA, whereas higher genetic predictions of the class Actinobacteria, class Lentisphaeria, family Bifidobacteriaceae, genus Blautia, order Bifidobacteriales, order Victivallales, and phylum Lentisphaerae were associated with lower risk. For the TAA subtype, on the basis of the IVW results, 13 gut microbial taxa were determined to have a causal relationship with TAA risk. Among them, the class Actinobacteria, family Bifidobacteriaceae, family Clostridiales vadin BB60 group, genus Coprococcus1, genus Veillonella, genus Intestinibacter, order Bifidobacteriales, order Victivallales, and phylum Lentisphaerae were associated with a lower TAA risk, whereas the family Oxalobacteraceae, family Pasteurellaceae, genus Eubacterium xylanophilum group, and genus Ruminococcaceae UCG014 were associated with higher TAA risk. Thirteen gut microbial taxa associated with AAA risk were also identified in the control and case groups (Supplementary Table 1). Among them, the class Lentisphaeria id.2250, family Victivallaceae id.2255, genus Lactococcus id.1851, genus Anaerotruncus id.2054, genus Blautia id.1992, and order Victivallales id.2254 were associated with lower AAA risk, whereas the genus Catenibacterium id.2153, genus Family XIII AD3011 group id.11293, genus Family XIII UCG001 id.11294, genus Oxalobacter id.2978, genus Bilophila id.3170, genus Escherichia Shigella id.3504, and genus Eubacterium brachy group id.11296 were associated with higher AAA risk. The relationship between the genus Family XIII AD3011 and AAA showed evidence of pleiotropy according to the MR-Egger method but no pleiotropy according to the MR-PRESSO method. On the basis of the overall results, the MR-PRESSO test was deemed reliable, indicating no horizontal pleiotropy. The above results are shown in Figures 2–4. Cochran’s Q test did not reveal significant heterogeneity (P > 0.05), and the MR-Egger and MR-PRESSO tests did not identify significant horizontal pleiotropy or outliers (P > 0.05) (Supplementary Table 1). The MR Steiger test showed no evidence of reverse causality (Supplementary Table 2).

Figure 2

Causal Effects of the Gut Microbiota on Aortic Aneurysm.

Figure 3

Causal Effects of the Gut Microbiota on Thoracic Aortic Aneurysm.

Figure 4

Causal Effects of the Gut Microbiota on Abdominal Aortic Aneurysm.

Bonferroni-Corrected Test

After Bonferroni correction, the results (Table 1) showed strong causal relationships between certain microbial taxa and aneurysms.

Table 1

MR Analysis Results of the Causal Relationship between the gut microbiota and AA.

Outcome	Classification	Nsnp	Methods	OR (95% CI)	P Value
AA	Phylum Lentisphaerae	9	Inverse variance weighted	0.82 (0.73–0.92)	0.001081677
			MR Egger	0.73 (0.45–1.18)	0.240365964
			Weighted median	0.85 (0.73–0.98)	0.030212552
			Weighted mode	0.88 (0.72–1.09)	0.275066916
	Class Lentisphaeria	8	Inverse variance weighted	0.81 (0.71–0.93)	0.002812266
			MR Egger	0.73 (0.44–1.22)	0.272388475
			Weighted median	0.83 (0.69–0.99)	0.035462053
			Weighted mode	0.89 (0.71–1.12)	0.352134357
	Family Bifidobacte\riaceae	11	Inverse variance weighted	0.79 (0.70–0.89)	8.00528 × 10−5
			MR Egger	0.93 (0.55–1.57)	0.787451213
			Weighted median	0.79 (0.63–0.98)	0.034571324
			Weighted mode	0.78 (0.59–1.04)	0.118112729
	Genus Family XIII UCG001	8	Inverse variance weighted	1.33 (1.14–1.55)	0.000203273
			MR Egger	1.10 (0.61–2.00)	0.765969749
			Weighted median	1.46 (1.12–1.91)	0.005139698
			Weighted mode	1.47 (0.99–2.20)	0.10001866
TAA	Phylum Lentisphaerae	9	Inverse variance weighted	0.81 (0.70–0.93)	0.002494393
			MR Egger	0.65 (0.35–1.21)	0.217125199
			Weighted median	0.82 (0.65–1.03)	0.084715554
			Weighted mode	0.96 (0.68–1.35)	0.808702009
	Family Bifidobacteriaceae	11	Inverse variance weighted	0.71 (0.61–0.82)	1.91475 × 10−6
			MR Egger	0.94 (0.44–1.99)	0.870227065
			Weighted median	0.70 (0.51–0.96)	0.025209257
			Weighted mode	0.68 (0.46–1.01)	0.082613371
	Genus Ruminococcaceae UCG014	10	Inverse variance weighted	1.30 (1.13–1.49)	0.000255698
			MR Egger	1.14 (0.66–1.98)	0.645934749
			Weighted median	1.17 (0.85–1.61)	0.324232843
			Weighted mode	1.14 (0.78–1.66)	0.52566428
AAA	Genus Bilophila	13	Inverse variance weighted	1.36 (1.17–1.57)	0.0000452
			MR Egger	1.55 (0.48–5.03)	0.483243703
			Weighted median	1.34 (0.98–1.85)	0.068775942
			Weighted mode	1.29 (0.80–2.08)	0.309338759

In AA, lower levels of the class Lentisphaeria (0.81, 95% CI: 0.71–0.93, P < 0.003), family Bifidobacteriaceae (0.79, 95% CI: 0.70–0.89, P < 0.0016), and phylum Lentisphaerae (0.82, 95% CI: 0.73–0.92, P < 0.005), as well as higher levels of the genus Family XIII UCG001 (1.33, 95% CI: 1.14–1.55, P < 0.0016), were strongly associated with AA. For the genus Blautia, because the IV contained only two SNPs, further sensitivity analysis could not be conducted; therefore, this genus was not included in the final results. Using leave-one-out analysis to sequentially remove SNPs, the predicted results for the remaining SNPs remain stable, thus indicating that the causal relationship was not driven by specific SNPs (Supplementary Figures 1–4).

In TAA, lower levels of the family Bifidobacteriaceae id.433 (0.71, 95% CI: 0.61–0.82, P < 0.0016) and phylum Lentisphaerae id.2238 (0.81, 95% CI: 0.70–0.93, P < 0.005), and higher levels of the genus Ruminococcaceae UCG014 id.11371 (1.30, 95% CI: 1.13–1.49, P < 0.0004), were strongly associated with TAA. Leave-one-out analysis did not show significant changes in the risk estimations (Supplementary Figures 5–7).

In AAA, higher levels of the genus Bilophila id.3170 (1.36, 95% CI: 1.17–1.57, P < 0.0004) were closely associated with AAA incidence. All SNP-specific F-statistics were greater than 10 (Supplementary Table 3), and MR Steiger tests (Supplementary Table 2) indicated no evidence of reverse causation. The forest plot results are shown in Supplementary Figure 8. The leave-one-out analysis also yielded stable results after iterative exclusion of each SNP (Supplementary Figure 8).

Reverse Analysis

The reverse causation analysis results (Supplementary Table 4) indicated that only AA might lead to higher levels of the genus Subdoligranulum (OR = 1.07, 95% CI: 1.03–1.11, P < 0.001) whereas the other results did not show statistical significance (P > 0.05).

Discussion

Previous studies on the relationship between AA and the gut microbiota were conducted primarily in animal models or prospective clinical trials [10, 11, 27]. However, prospective randomized trials are time-consuming and costly, and confounding factors are difficult to control. In such cases, genotypes can be considered to replace exposure factors for causal inference, in a process akin to a natural randomized controlled trial. A recent study has explored the relationship between AAA and the gut microbiota through MR and arrived at conclusions similar to ours. Therefore, we further extended our investigation to explore the relationship between AA and its subtypes, including TAA and AAA, with the gut microbiota. Additionally, we conducted Bonferroni correction to enhance the credibility of the results [28]. In this study, we used MR to analyze the potential causal relationship between AA and its subtypes and the gut microbiota at the genetic prediction level. Our results were based on a large GWAS database, to achieve reliable causal interpretation effects. Our findings may provide guidance for future AA treatment targeting specific gut bacteria in a genomic context.

We identified 26 microbial communities associated with the incidence of AA and its subtypes, six of which showed strong causal relationships. Among them, the genetic prediction level of the phylum Lentisphaerae may be associated with diminished risk of AA and TAA. In contrast, the class Lentisphaeria, a member of the phylum Lentisphaerae, was associated with a lower incidence of only AA but was associated with a lower risk of AAA before Bonferroni correction. The phylum Lentisphaerae, a member of the Planctomycetes-Verrucomicrobia-Chlamydia superphylum, is currently divided into two classes: Lentisphaeria and Oligosphaeria [29, 30]. Lentisphaeria includes two orders (Lentisphaerales and Victivallales) and two families (Lentisphaeraceae and Victivallaceae). Previous studies have suggested that Lentisphaerae bacteria promote polysaccharide degradation, facilitate phosphorus and iron metabolism, and increase and maintain microbial diversity [31]. Studies have found that Lentisphaerae dysbiosis is associated with ileitis [32], autoimmune hepatitis [33], and systemic sclerosis [34], through mechanisms potentially associated with inflammation, whereas the inflammatory response may promote the progression of AA through pathways such as atherosclerosis. However, because Lentisphaerae members are difficult to culture, the above conclusions are based on the results of metagenomic predictions, and further experimental testing or clinical validation is necessary in the future to confirm the cellular biology functions of Lentisphaerae bacteria and their role in the occurrence of AA.

Herein, higher levels of the gene-predicted family Bifidobacteriaceae were associated with lower risks of AA and TAA. Bifidobacterium is an anaerobic, bile-resistant, Gram-negative rod-shaped bacterium [35]. Studies have shown that Bifidobacterium may directly regulate intestinal function by modulating the expression of genes associated with nutrient absorption, mucosal barrier enhancement, and vascular endothelial growth factor expression [36]. AA is relatively more common in older people [37], whose gut microbiota is characterized by diminished Bacteroides spp., Bifidobacterium spp., and Lactobacilli spp. Chiara et al. [38] have found that older people who consume probiotics have elevated levels of the family Bifidobacteriaceae in their gut microbiota, along with elevated total short-chain fatty acids (SCFAs) and butyric acid levels in fecal samples, and diminished serum hsCRP. SCFAs include primarily acetate, butyrate, and propionate, which are metabolites of the intestinal microbiota, and have functions in regulating the immune system and maintaining the intestinal epithelial barrier [39]. Therefore, the family Bifidobacteriaceae may suppress the occurrence of AA and TAA by maintaining intestinal function and exerting anti-inflammatory effects.

The genera Blautia and Ruminococcaceae UCG014 both belong to the phylum Bacillota. Interestingly, although further sensitivity analysis could not be conducted, our results nonetheless suggested that the gene-predicted level of the genus Blautia negatively correlated with the occurrence of AA and AAA, whereas that of the genus Ruminococcaceae UCG014 positively correlated with TAA occurrence. Blautia is a symbiotic strict anaerobe that has been reported to prevent inflammation and maintain intestinal balance environment through the production of SCFAs and upregulation of regulatory T cells in the gut [40]. The abundance of Blautia is diminished in patients with inflammatory bowel disease [41] and colorectal cancer [42]. Thus, Blautia is a probiotic with positive effects on human health [43]. The genus Ruminococcaceae UCG014 also has a role in producing SCFAs [44]; however, our study suggested that it is associated with a high risk of TAA; further research may be necessary to confirm this finding.

In addition, our results indicated that the genus Bilophila and genus Family XIII UCG001 were significantly associated with elevated risk of AAA and AA, respectively. Bilophila species have been shown to promote intestinal barrier dysfunction, bile acid metabolism disorders, and inflammation in mouse models [45]. In vitro experiments have shown that Bilophila species can convert taurine to hydrogen sulfide, which may play an important role in systemic inflammation [46]. However, the relationship between genus Family XIII UCG001 and diseases has not been clearly demonstrated. In an experiment conducted using mice as experimental subjects, the results suggested that the abundance of genus Family XIII UCG001 may be associated with arachidonic acid [47], and arachidonic acid and its metabolites have been associated with various cardiovascular diseases [48]. Both these genera may affect the development of AA or AAA through inflammatory responses.

Our study has several limitations. First, the gut microbiota exhibits high heterogeneity and interindividual variability, thus significantly decreasing the statistical power of microbiome GWAS analysis and potentially leading to pleiotropy. Therefore, we counted only IVs with F statistics greater than 10 and used various sensitivity analyses to decrease the effects of weak instrument bias and multicollinearity variability. Second, MR analysis is based on untestable assumptions, and further experimental and clinical validation studies are required to assess the clinical significance of microbial species. Furthermore, because of the lack of population demographic data, we were unable to perform subgroup analyses, such as investigation of the causal relationship between sex- and age-specific gut microbiota and AA. Finally, as described earlier, variations in the incidence of AAA occur among countries, and populations in different geographical regions also exhibit distinct gut microbiota characteristics [49]. Subsequent validation with GWAS data from more countries is necessary.

Conclusion

The occurrence of AA and its subtypes is closely associated with the gut microbiota. Reverse causal analysis did not indicate a significant influence in the presence of gut microbiota that promote or inhibit AA and its subtypes. Interestingly, the gut microbiota affecting the occurrence of AAA or TAA do not overlap, thus potentially further indicating that AAA and TAA have different pathogenic mechanisms. We may further investigate these specific microbial communities to explore their roles in AA and its subtypes. Our findings may potentially provide guidance for the development of potential therapeutic targets for AA and its subtypes from the perspective of the gut microbiota. In summary, we used gene prediction to determine specific gut microbiota that might serve as useful biomarkers or therapeutic targets for early disease diagnosis or prevention.

Data Availability Statement

Datasets for this study can be found in the FinnGen database (https://www.finngen.fi/en) and MiBioGen (https://mibiogen.gcc.rug.nl/menu/main/home).

Ethics Statement

Ethical approval was not provided for this study, which used solely publicly available aggregated data and did not involve direct interaction or interventions with human participants.

Author Contributions

Yan Lv and Dexin Shen performed the studies, participated in data collection, performed the statistical analysis and drafted the manuscript, under the supervision of Jinying Zhang and Junnan Tang. Ge Zhang, Bo Wang and Haiyu Wang participated in data collection and proofread the manuscript. All authors read and approved the final manuscript.

Acknowledgements

We thank researchers in the MiBioGen Consortium and the FinnGen project for their excellent work and sharing of data.

Conflicts of Interest

The authors declare no conflicts of interest.

Supplementary Material

Supplementary material for this paper is available at the following links.

Supplementary table information: https://cvia-journal.org/wp-content/uploads/2024/03/CVIA_411_SUPPLEMENTARY_TABLES.pdf

Supplementary tables can be downloaded here: https://cvia-journal.org/wp-content/uploads/2024/03/Supplementary_tables_1_5.xlsx

Supplementary figures can be downloaded at the following links.

https://cvia-journal.org/wp-content/uploads/2024/03/Supplementary_FIG_S1.pdf

https://cvia-journal.org/wp-content/uploads/2024/03/Supplementary_FIG_S2.pdf

https://cvia-journal.org/wp-content/uploads/2024/03/Supplementary_FIG_S3.pdf

https://cvia-journal.org/wp-content/uploads/2024/03/Supplementary_FIG_S4.pdf

https://cvia-journal.org/wp-content/uploads/2024/03/Supplementary_FIG_S5.pdf

https://cvia-journal.org/wp-content/uploads/2024/03/Supplementary_FIG_S6.pdf

https://cvia-journal.org/wp-content/uploads/2024/03/Supplementary_FIG_S7.pdf

https://cvia-journal.org/wp-content/uploads/2024/03/Supplementary_FIG_S8.pdf

Citation Information

Download Citation.

References

1. KW Johnston, RB Rutherford, MD Tilson, DM Shah, L Hollier, JC Stanley. Suggested standards for reporting on arterial aneurysms. Subcommittee on reporting standards for arterial aneurysms, ad hoc committee on reporting standards, Society for vascular surgery and North American Chapter, International Society for Cardiovascular Surgery. J Vasc Surg 1991;13(3):452–8.

2. RA Quintana, WR Taylor. Introduction to the compendium on aortic aneurysms. Circ Res 2019;124(4):470–1.

3. RA Quintana, WR Taylor. Cellular mechanisms of aortic aneurysm formation. Circ Res 2019;124(4):607–18.

4. G Zhang, Z Liu, J Deng, L Liu, Y Li, S Weng, et al. Smooth muscle cell fate decisions decipher a high-resolution heterogeneity within atherosclerosis molecular subtypes. J Transl Med 2022;20(1):568.

5. PW Stather, N Dattani, MJ Bown, JJ Earnshaw, TA Lees. International variations in AAA screening. Eur J Vasc Endovasc Surg 2013;45(3):231–4.

6. JF Blanchard, HK Armenian, PP Friesen. Risk factors for abdominal aortic aneurysm: results of a case-control study. Am J Epidemiol 2000;151(6):575–83.

7. H Liu, X Chen, X Hu, H Niu, R Tian, H Wang, et al. Alterations in the gut microbiome and metabolism with coronary artery disease severity. Microbiome 2019;7(1):68.

8. T Yang, MM Santisteban, V Rodriguez, E Li, N Ahmari, JM Carvajal, et al. Gut dysbiosis is linked to hypertension. Hypertension 2015;65(6):1331–40.

9. P-C Chen, Y-W Chien, S-C Yang. The alteration of gut microbiota in newly diagnosed type 2 diabetic patients. Nutrition 2019;63–64:51–6.

10. TW Benson, KA Conrad, XS Li, Z Wang, RN Helsley, RC Schugar, et al. Gut microbiota-derived trimethylamine N-oxide contributes to abdominal aortic aneurysm through inflammatory and apoptotic mechanisms. Circulation 2023;147(14):1079–96.

11. Z Tian, Y Zhang, Z Zheng, M Zhang, T Zhang, J Jin, et al. Gut microbiome dysbiosis contributes to abdominal aortic aneurysm by promoting neutrophil extracellular trap formation. Cell Host Microbe 2022;30(10):1450–63.e8.

12. H Li, H Xu, Y Li, Y Jiang, Y Hu, T Liu, et al. Alterations of gut microbiota contribute to the progression of unruptured intracranial aneurysms. Nat Commun 2020;11(1):3218.

13. F Shikata, K Shimada, H Sato, T Ikedo, A Kuwabara, H Furukawa, et al. Potential influences of gut microbiota on the formation of intracranial aneurysm. Hypertension 2019;73(2):491–6.

14. R Shinohara, H Nakashima, T Emoto, T Yamashita, Y Saito, N Yoshida, et al. Gut microbiota influence the development of abdominal aortic aneurysm by suppressing macrophage accumulation in mice. Hypertension 2022;79(12):2821–9.

15. G Davey Smith, G Hemani. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet 2014;23(R1):R89–98.

16. A Kurilshikov, C Medina-Gomez, R Bacigalupe, D Radjabzadeh, J Wang, A Demirkan, et al. Large-scale association analyses identify host factors influencing human gut microbiome composition. Nat Genet 2021;53(2):156–65.

17. KJ van der Velde, F Imhann, B Charbon, C Pang, D van Enckevort, M Slofstra, et al. MOLGENIS research: advanced bioinformatics data software for non-bioinformaticians. Bioinformatics 2019;35(6):1076–8.

18. S Burgess, SG Thompson; CRP CHD Genetics Collaboration. Avoiding bias from weak instruments in Mendelian randomization studies. Int J Epidemiol 2011;40(3):755–64.

19. JR Staley, J Blackshaw, MA Kamat, S Ellis, P Surendran, BB Sun, et al. PhenoScanner: a database of human genotype-phenotype associations. Bioinformatics 2016;32(20):3207–9.

20. MI Kurki, J Karjalainen, P Palta, TP Sipilä, K Kristiansson, KM Donner, et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature 2023;613(7944):508–18.

21. J Bowden, G Davey Smith, S Burgess. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol 2015;44(2):512–25.

22. J Bowden, G Davey Smith, PC Haycock, S Burgess. Consistent estimation in mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol 2016;40(4):304–14.

23. J Bowden, MF Del Greco, C Minelli, G Davey Smith, N Sheehan, J Thompson. A framework for the investigation of pleiotropy in two-sample summary data Mendelian randomization. Stat Med 2017;36(11):1783–802.

24. M Verbanck, C-Y Chen, B Neale, R Do. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet 2018;50(5):693–8.

25. G Hemani, K Tilling, G Davey Smith. Orienting the causal relationship between imprecisely measured traits using GWAS summary data. PLoS Genet 2017;13(11):e1007081.

26. VW Skrivankova, RC Richmond, BAR Woolf, J Yarmolinsky, NM Davies, SA Swanson, et al. Strengthening the reporting of observational studies in epidemiology using mendelian randomization: the STROBE-MR statement. JAMA 2021;326(16):1614–21.

27. S Liu, Y Liu, J Zhao, P Yang, W Wang, M Liao. Effects of spermidine on gut microbiota modulation in experimental abdominal aortic aneurysm mice. Nutrients 2022;14(16):3349.

28. X Zhou, W Ruan, T Wang, H Liu, L Du, J Huang. Exploring the impact of gut microbiota on abdominal aortic aneurysm risk through a bidirectional Mendelian randomization analysis. J Vasc Surg 2023.

29. A Choi, J Song, Y Joung, K Kogure, J-C Cho. Lentisphaera profundi sp. nov., isolated from deep-sea water. Int J Syst Evol Microbiol 2015;65(11):4186–90.

30. Y-L Qiu, M Muramatsu, S Hanada, Y Kamagata, R-B Guo, Y Sekiguchi. Oligosphaera ethanolica gen. nov., sp. nov., an anaerobic, carbohydrate-fermenting bacterium isolated from methanogenic sludge, and description of Oligosphaeria classis nov. in the phylum Lentisphaerae. Int J Syst Evol Microbiol 2013;63(Pt 2):533–9.

31. T Zhang, R Zheng, R Liu, R Li, C Sun. Cultivation and functional characterization of a deep-sea lentisphaerae representative reveals its unique physiology and ecology. Front Mar Sci 2022;9:848136.

32. HN Fan, P Zhu, YM Lu, JH Guo, J Zhang, GQ Qu, et al. Mild changes in the mucosal microbiome during terminal ileum inflammation. Microb Pathog 2020;142:104104.

33. J Lou, Y Jiang, B Rao, A Li, S Ding, H Yan, et al. Fecal microbiomes distinguish patients with autoimmune hepatitis from healthy individuals. Front Cell Infect Microbiol 2020;10:342.

34. E Russo, S Bellando-Randone, D Carboni, BS Fioretto, E Romano, S Baldi, et al. The differential crosstalk of the skin-gut microbiome axis as a new emerging actor in systemic sclerosis. Rheumatology (Oxford) 2024;63:226–34.

35. HM Wexler. Bacteroides: the good, the bad, and the nitty-gritty. Clin Microbiol Rev 2007;20(4):593–621.

36. M Freitas, E Tavan, C Cayuela, L Diop, C Sapin, G Trugnan. Host-pathogens cross-talk. Indigenous bacteria and probiotics also play the game. Biol Cell 2003;95(8):503–6.

37. T Odamaki, K Kato, H Sugahara, N Hashikura, S Takahashi, JZ Xiao, et al. Age-related changes in gut microbiota composition from newborn to centenarian: a cross-sectional study. BMC Microbiol 2016;16:90.

38. C Salvesi, S Silvi, D Fiorini, S Scortichini, G Sagratini, FA Palermo, et al. Impact of a probiotic diet on well-being of healthy senior: THE PROBIOSENIOR PROJECT. J Appl Microbiol 2022;133(5):2941–53.

39. H Mäkivuokko, K Tiihonen, S Tynkkynen, L Paulin, N Rautonen. The effect of age and non-steroidal anti-inflammatory drugs on human intestinal microbiota composition. Br J Nutr 2010;103(2):227–34.

40. CH Kim, J Park, M Kim. Gut microbiota-derived short-chain Fatty acids, T cells, and inflammation. Immune Netw 2014;14(6):277–88.

41. L Chen, W Wang, R Zhou, SC Ng, J Li, M Huang, et al. Characteristics of fecal and mucosa-associated microbiota in Chinese patients with inflammatory bowel disease. Medicine 2014;93(8):e51.

42. ML Richard, G Liguori, B Lamas, G Brandi, G da Costa, TW Hoffmann, et al. Mucosa-associated microbiota dysbiosis in colitis associated cancer. Gut Microbes 2018;9(2):131–42.

43. X Liu, B Mao, J Gu, J Wu, S Cui, G Wang, et al. Blautia-a new functional genus with potential probiotic properties? Gut Microbes 2021;13(1):1–21.

44. B Tian, J Zhao, M Zhang, Z Chen, Q Ma, H Liu, et al. Lycium ruthenicum anthocyanins attenuate high-fat diet-induced colonic barrier dysfunction and inflammation in mice by modulating the gut microbiota. Mol Nutr Food Res 2021;65(8):e2000745.

45. JM Natividad, B Lamas, HP Pham, ML Michel, D Rainteau, C Bridonneau, et al. Bilophila wadsworthia aggravates high fat diet induced metabolic dysfunctions in mice. Nat Commun 2018;9(1):2802.

46. SC Peck, K Denger, A Burrichter, SM Irwin, EP Balskus, D Schleheck. A glycyl radical enzyme enables hydrogen sulfide production by the human intestinal bacterium Bilophila wadsworthia. Proc Natl Acad Sci U S A 2019;116(8):3171–6.

47. Y Li, C Liu, J Luo, Y Zeng, X Meng, S Wang, et al. Ershiwuwei Lvxue Pill alleviates rheumatoid arthritis by different pathways and produces changes in the gut microbiota. Phytomedicine 2022;107:154462.

48. T Sonnweber, A Pizzini, M Nairz, G Weiss, I Tancevski. Arachidonic acid metabolites in cardiovascular and metabolic diseases. Int J Mol Sci 2018;19:3285.

49. AM Porras, Q Shi, H Zhou, R Callahan, G Montenegro-Bethancourt, N Solomons, et al. Geographic differences in gut microbiota composition impact susceptibility to enteric infection. Cell Rep 2021;36(4):109457.