Deciphering the Role of Metformin in Abdominal Aortic Aneurysm Progression: Insights from Two-Step Mendelian Randomization and Colocalization Analysis

Deciphering the Role of Metformin in Abdominal Aortic Aneurysm Progression: Insights from Two-Step Mendelian Randomization and Colocalization Analysis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Deciphering the Role of Metformin in Abdominal Aortic Aneurysm Progression: Insights from Two-Step Mendelian Randomization and Colocalization Analysis Zhaoxuan Zhang, Yuemeng Li, Jian Wang, Xiaoxu Zhang, Deying Jiang, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6834737/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Objective: Our research sought to investigate the relationship between metformin and abdominal aortic aneurysm (AAA) risk. We aimed to contribute to the current understanding of metformin's potential as a pharmaceutical intervention for AAA and explore the genetic facets influencing its effect. Methods: The methods encompassed meta-analysis, Mendelian Randomization (MR) and colocalization analysis. Meta-analysis facilitated a robust review and synthesis of the current literature. MR could investigate the causal relationship between metformin treatment and AAA incidence, and the mediating effects of certain factors. Colocalization analysis results are used as a supplement to MR analysis. Results: Meta-analysis suggest that metformin prescription is associated with a clinically important significant reduction in both growth in people with AAA. MR analysis results indicated the negative correlation between metformin treatment and AAA (OR = 0.0108, 95% CI: 0.000204 - 0.572, P = 0.0254), and the reduction of total cholesterol levels (mediation effect 25.17%, OR=0.3199, 95% CI: 0.159 - 0.644) mediated the protective effect of metformin against AAA, and the main role may be played by low-density lipoprotein levels (mediation effect 28.84%, OR = 0.271, 95% CI: 0.133 - 0.552). Colocalization analysis identified 9 significant genes, notably GPD2, associated with both metformin treatment and reduced AAA risk. Conclusions: Metformin treatment is associated with a significant reduction in AAA incidence, potentially mediated by its effects on blood lipid levels. These findings support further investigation into metformin as a therapeutic option for AAA, emphasizing the importance of integrating pharmacological and genetic approaches in cardiovascular disease management. Abdominal Aortic Aneurysm Metformin Mendelian Randomization Pharmacotherapy Therapeutic Targets Molecular Mechanisms Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 1. Introduction Understanding the multifaceted role Interactions of biological factors play in the development and progression of diseases is a fundamental aspect of biomedical research [ 1 ] . An aspect that remains underexplored is the impact and relationship between diabetes treatment, particularly metformin, and cardiovascular disease. A specific area of focus in cardiovascular disorders is Abdominal Aortic Aneurysm (AAA), a condition of multifactorial etiology characterized by a progressive enlargement of the aorta, leading to aortic rupture if left undetected or untreated [ 2 ] . Abdominal aortic aneurysm (AAA) is a grave life-threatening cardiovascular disorder characterized by abnormal dilation of the aorta. This rupture-prone condition often remains asymptomatic until there is a rupture, leading to internal bleeding and potential fatality. The lack of specific symptoms poses a significant challenge in timely diagnosis and effective disease management, signaling an unmet need for novel detection and treatment strategies. Currently, the therapeutic interventions for AAA are majorly restricted to surgical approaches like open repair and endovascular aneurysm repair 3 . The pharmaceutical landscape for AAA is unfortunately sparse. Some recommendations include beta-blockers [ 4 ] , statins [ 5 ] , antiplatelet agents, and angiotensin-converting enzyme (ACE) inhibitors [ 6 ] which primarily focus on managing the risk factors, such as hypertension and hyperlipidemia that contribute to AAA incidence. However, their efficacy in hindering AAA growth and rupture remains inconclusive, warranting more research. Metformin, the first-line therapy for type II diabetes, has been pivotal in the control and management of the disease, contributing immensely to therapeutic regimens due to its commendable safety profile, and high efficacy. Still, further illumination on its impact in the arena of cardiovascular health, though well established, remains in the early exploration phases [ 7 ] . The promising protective roles of metformin in cardiovascular health open avenues and pique both clinical and academic interest in further determining its potential as a useful tool in fighting cardiovascular diseases [ 8 ] . Appreciating the medicine's impact beyond its primary role as an oral hypoglycemic agent, addressing insulin resistance, might yield novel therapeutic and prognostic pathways in other diseases. However, the literature concerning its effectiveness in AAA treatment has shown mixed results, opening gates for further investigation [ 9 ] . A multitude of studies, conducted primarily in rodent models, have delved into the impact of metformin on AAA [ 10 – 14 ] . While these studies have shed light onto the possible effects, they have their limitations. Firstly, rodent models, although scientifically valuable, do not perfectly mimic the human pathology in terms of anatomy, physiology, and disease mechanisms. Therefore, the translational potential of these findings to humans might be restricted, thus calling for caution in interpretation. Secondly, another barrier is the dearth of Randomized Controlled Trials (RCTs). RCTs, often considered the gold standard in clinical research, provide powerful and reliable evidence on the efficacy of interventions. Given their experimental nature, RCTs often yield high-quality outcomes and are less prone to bias [ 15 ] . They allow the demonstration of the cause and effect, providing robust evidence on efficacy and effectiveness. However, there remains a dearth of such trials intending to investigate the therapeutic potential of drugs like metformin in AAA management. Various factors, including resource constraints and the complexity of disease mechanisms, contribute to the challenges inhibiting the execution of RCTs. Enter Mendelian randomization (MR), an analytical method using genetic variants as instrumental variables [ 16 ] . It has been proposed as a solution to circumnavigate the above-mentioned challenges. By utilizing genetic variants that are associated with an exposure (in our case, metformin treatment), MR studies can help estimate the causal effect of the exposure on the outcome (AAA in this context). By harnessing the principles of genetics, where the alleles of genetic variants are randomly assigned at conception, MR studies mimic the randomization process naturally occurring in RCTs [ 17 ] . In other words, Mendelian randomization can estimate the causal effect of specific modifiable exposures on disease outcomes, providing evidence that can complement or, in some cases, substitute the need for performing RCTs [ 18 ] . This methodology is particularly useful for investigating questions that may be challenging or impossible to address with RCTs due to ethical or practical reasons. Its application can, therefore, enable us to scrutinize the impact of metformin in AAA, thereby filling the gap left by the absence of RCTs. Our study is nested in this research gap, intent on exploring the impact of metformin therapy on the progression and clinical events related to AAA, particularly in patients without diabetes. The objective is to dissect the relationship that exists between metformin and AAA, to establish a probable causative effect, if any. This step is crucial as it forms the background on which we can build our understanding of the magnitude of the impact and subsequently design interventions. Safety, quality of life, and overall survival of patients may hinge partly on this valuable knowledge. By embarking on our study with MR, we aim to delve deeper into the complex interplay between AAA and metformin, uncovering insights that may potentially revolutionize therapeutic approaches to AAA. The limitations and advantages, findings and interpretations, and implications for future research will be discussed in detail in the sections to follow. 2. Material and methods 2.1 Meta-analysis for association between metformin and abdominal aortic aneurysm 2.1.1 Literature Search Strategy In accordance with the PRISMA guidelines, an exhaustive literature search was executed across Medline, Embase, Cochrane Library, and Ovid databases, covering a timeframe from March 1, 1999 to November 11, 2023 [19] . The search was centered around the key term "aortic aneurysm, abdominal". To ensure a comprehensive coverage, the search was broadened to include an array of synonyms and related terms for Metformin, such as Dimethylbiguanidine, Dimethylguanylguanidine, Glucophage, Metformin Hydrochloride, and others. The search strategy also encompassed additional terms such as "Aortic aneurysm (Aneurysms, Aortic, Aortic Aneurysms, Aneurysm, Aortic)," "Aortic dilatation," to ensure all relevant studies were captured. To eliminate any bias and ensure objectivity, two independent reviewers were engaged to screen every article and select cohort studies for inclusion. This comprehensive search was not limited by language, ensuring a global perspective in the selection of studies. The resources scrutinized for this analysis spanned published articles, conference abstracts accompanied by statistical data and charts, as well as pertinent literature cited in the reference lists of the included articles. In cases where additional data was deemed necessary for a more comprehensive analysis, authors were contacted directly. We ensured that every piece of relevant information was captured, thereby enhancing the robustness and reliability of our meta-analysis. 2.1.2 Eligibility Criteria The eligibility of the articles was determined by two independent reviewers. Any conflicts between the reviewers were resolved by consulting a third independent reviewer. The studies considered for inclusion were randomized controlled trials, cohort studies, and case-control studies that satisfied the following criteria: The studies focused on AAA patients who were on Metformin’s medication, and they reported the AAA growth rate. Conference abstracts were included if they provided sufficient data for analysis. We deemed studies as eligible if they were cohort studies involving adult diabetic patients who had undergone a minimum of 8 weeks of metformin pharmacologic intervention. We excluded (1) reviews, case reports, comments, recommendations, letters, ongoing trials, protocols, conference abstracts, consensus or statements, and articles that lacked applicable data; (2) duplicate reports and studies of low quality, inconsistent type, or those providing insufficient information; and (3) studies using inappropriate statistical methods or providing insufficient data. The selected articles needed to clearly demonstrate the effect of metformin pharmacologic therapy on changes in aortic aneurysm diameter or on events related to aortic dilation as outcome results. 2.1.3 Data Extraction The data extraction process was carried out by two independent researchers who sought out pertinent articles, screened potential ones based on the set eligibility criteria, and performed data extraction using a standardized datasheet independently. Any disagreement was addressed through discussions involving a third researcher. The extracted data covered the following details: authorship, publication year, country of origin, study type, inclusion criteria, participant numbers, interventions (specifically, metformin), and study results. Treatment strategies for aortic changes in patients with aortic aneurysms, with or without the use of metformin, were compared without any restrictions on treatment history. The main data points extracted included: the first author, year of publication, study design, initial AAA diameter, sample size, follow-up duration, and AAA growth rate. The methodological quality of cohort and case-control studies was evaluated using the Newcastle-Ottawa scale (NOS) [ 2 0] . The two researchers independently completed data extraction and quality assessment, and any discrepancies were resolved through discussion. 2.1.4 Quality Assessment To assess the quality of the studies, we used the Newcastle-Ottawa Scale (NOS). The NOS checklist has three quality parameters: (1) selection of the study groups; (2) comparability of the groups; and (3) ascertainment of either the exposure or outcome of interest for case-control or cohort studies respectively. Each study received a score ranging from zero to nine. Studies that achieved a score of seven or more were deemed to be of high quality. 2.1.5 Statistical Analysis The statistical analysis was conducted using Review Manager (version 5.4). A meta-analysis was performed to compare AAA growth between the intervention and control groups. The AAA growth (mm/year) in both groups was presented as a mean and standard deviation (SD), and the results were expressed as a mean difference (MD) with its corresponding 95% confidence interval (95% CI) [ 2 1] . Heterogeneity across the studies was quantitatively assessed by the I 2 statistic, with a threshold of 50% indicating significant heterogeneity. If statistical heterogeneity was not present (I 2 50%), the random effects model was used, and a sensitivity analysis was performed to pinpoint the source of heterogeneity [ 2 2] . If the cause of heterogeneity could not be determined, we continued with the random-effects model [ 2 3] . Statistical significance was denoted by a P -value less than 0.05. The results of the analysis were visualized using forest plots. 2.1.6 Assessment of Publication Bias To evaluate the presence of publication bias in our meta-analysis, we employed funnel plots, a widely recognized method for detecting such biases. These plots are particularly effective in illustrating asymmetry, which often indicates potential publication bias. The funnel plot approach involves plotting the treatment effects estimated from individual studies against a measure of study size or precision, usually the standard error. We created funnel plots for each outcome measure analyzed in our meta-analysis. Visual inspection of these plots was the first step in assessing publication bias. A symmetrical distribution of studies within the funnel plot suggested the absence of publication bias, while asymmetry indicated its potential presence. Asymmetry in funnel plots can also result from other factors such as heterogeneity among study methodologies or true differences in effect sizes. Therefore, the results of the funnel plot analysis should be interpreted with caution, considering other contextual factors related to the studies included in the meta-analysis. 2.2 Mendelian randomization analysis 2.2.1 Study Design The design of this study unfolds as follows, and is graphically represented in Figure 1 . We utilized summary-level data extracted from Genome-Wide Association Studies (GWAS) to form the foundation of our analysis. Our primary focus in this study is on five genes that are known to be targeted by metformin, a medication widely used in the treatment of type 2 diabetes. These genes, namely MCI, AMPK, GDF15, MG3, and FBP1, were identified and selected based on extensive reviews of existing literature [24, 25] . Our aim was to investigate whether there is a direct causal relationship between metformin treatment, the effects of these specific genes, and the incidence of abdominal aortic aneurysm (AAA), and to explore the mediating role of serum lipids, including high-density lipoprotein (HDL), low-density lipoprotein (LDL), triglycerides (TG) and total cholesterol (TC), between metformin treatment and AAA. The mediation effect analysis was completed using the traditional two-step method, and the indirect effect and proportion were obtained using the delta method [26] . MR is a method that uses genetic variants, or single nucleotide polymorphisms (SNPs), as instrumental variables (IVs) to estimate the causal effect of an exposure (in this case, metformin treatment and its target genes) on an outcome (AAA incidence). This approach is particularly valuable as it helps to overcome the limitations of conventional observational studies, such as confounding and reverse causation. MR analysis is built on three core assumptions. Firstly, we assume that the SNPs we select as IVs are strongly associated with the exposure. This means that these genetic variants should directly influence the action of metformin or its target genes. Secondly, these SNPs must be independent of any confounding factors that might distort the true relationship between exposure and outcome. This is to ensure that any observed associations are not due to these confounding variables, but rather are indicative of a direct causal link. Finally, the third assumption is that the IVs are related to the outcome only via the exposure. In other words, these SNPs should influence AAA incidence only through their effect on metformin treatment and its target genes, and not via any other biological pathways. This assumption is critical to maintain the validity of our MR analysis. 2.2.2 Data sources The data used in this study is drawn entirely from the Integrated Exposure Unit (IEU) Open GWAS project, FinRegistry project and Ensembl genome database project. IEU Open GWAS is a comprehensive summary database that contains information from a multitude of Genome-Wide Association Studies (GWAS). FinRegistry is a joint research project of the Finnish Institute of Health and Welfare (THL) and the Data Science and Genetic Epidemiology Lab research group at the Institute for Molecular Medicine Finland (FIMM), University of Helsinki. The specific characteristics of the datasets used in our study are presented in Table 1 . Our research operates under the ethical approvals granted to the original GWAS studies from which we sourced our data. This means we were not required to obtain new informed consent from patients or to meet any additional ethical requirements. All the datasets we selected to use were derived entirely from European population samples. The first set of data we used pertains to the exposure of individuals to metformin treatment. This data was obtained from the UK Biobank dataset, a resource that includes genetic and health information from a staggering total of 462,933 individuals. Within this dataset, there were 11,552 cases (individuals who had received metformin treatment) and 451,381 controls (those who had not received metformin treatment). Crucially, this classification was based solely on whether or not the individuals had been exposed to metformin, without any consideration of variables such as age or gender. To investigate the exposure of individuals to metformin's target genes, we turned to a different dataset from Ensembl, which includes data from 30,721 individual subjects. Ensembl is a comprehensive database providing detailed information about the genes, variants, and functional interpretations in a wide range of species. The dataset of lipid indexes as mediating factors came from a metabolomics study of 115,082 European individuals published in 2022 by T. G. Richardson et al. The outcome dataset, which includes information on the incidence of Abdominal Aortic Aneurysm (AAA), was sourced from the FinnGen dataset. This dataset includes a total of 473,681 individuals, with a breakdown of 4,217 cases (those diagnosed with AAA) and 469,464 controls (those without AAA). The standard definition code for AAA cases is I71.3 and I71.4 from the International Classification of Diseases 10th revision (ICD-10). 2.2.3 Selection of instrumental variables The process of selecting instrumental variables (IVs) for our study was involving multiple stages of filtering and testing to ensure the validity of our MR analysis. Our first step was to identify potential IVs among independent genetic variants that exhibited genome-wide significance, defined as having a P-value less than 5e-08. This indicates that there is a less than one in twenty million chance that the observed associations between these genetic variants and the exposure variables occurred by random chance. Thus, selecting SNPs with genome-wide significance adds a layer of robustness to our analysis. After potential IVs were identified, we proceeded to test for linkage disequilibrium (LD) among the selected single nucleotide polymorphisms (SNPs). LD is a phenomenon that occurs when SNPs are inherited together more often than would be expected by chance. This can introduce bias into our analysis, as it could artificially inflate the perceived association between our IVs and the exposure variables. To avoid this, we set stringent LD thresholds: for SNPs related to metformin treatment and phenotypes used as mediators in the two-step analysis, we only retained those with an r2 value less than 0.01 and a physical distance greater than 10000 kilobases (kb); for SNPs related to the target genes, we only retained those with an r2 value greater than 0.1 and a distance greater than 100kb. Next, we discarded any weak IVs, defined as those with an F-statistic less than 10. The F-statistic is a measure of the strength of the instrument, and a low F-statistic can suggest that the IV is weakly correlated with the exposure. In this study, the F statistic was calculated as follows: F = R2(n - k - 1)/k (1 - R2), where R2, n, and k denote the proportion of variance in exposure explained by selected genetic tools, the sample size of the exposure GWAS, and the number of chosen genetic tools, respectively [ 27 ] . Finally, we harmonized the palindromic SNPs to ensure that the alleles of the IVs in the exposure and outcome datasets had consistent effects. Palindromic SNPs are those where the major and minor alleles cannot be unambiguously determined due to their palindromic nature (A/T or C/G). Detailed information on all used SNPs can be found in Supplementary Tables 1-11 . 2.2.4 Statistical analysis We employed five commonly used Mendelian Randomization (MR) methods, each providing a unique approach to analyze the data. A significance threshold of P < 0.05 was established for all tests. A. Inverse-Variance Weighted (IVW) Method: This method was the primary MR approach we used to infer causal relationships. The IVW method operates under the assumption that all instrumental variables (IVs) exert a common causal effect on the outcome through the exposure. It aggregates the effect estimates from each IV, using a meta-analysis-like framework. In this framework, the inverse variance of each effect estimate is used as a weight, allowing us to derive a summarized causal estimate. This approach is particularly powerful when the IVs are strong and there is no evidence of horizontal pleiotropy [28, 29] . B. MR-Egger Method: We used the MR-Egger method specifically to test for horizontal pleiotropy, which occurs when IVs influence the outcome through pathways that are not related to the exposure. This is crucial because such pleiotropy can bias MR estimates. The MR-Egger intercept provides a test for directional pleiotropy; a P -value of less than 0.05 here suggests significant horizontal pleiotropy [30] . C. Weighted Median, Weighted Mode, and Simple Mode Methods: These methods offer alternative ways to calculate the causal estimate. They are particularly useful when there is potential invalidity among some IVs, as they can provide more robust estimates under certain conditions of IV invalidity [31 , 3 2] . D. Cochran’s Q Test for Heterogeneity: To assess the heterogeneity among the IV estimates, we employed Cochran’s Q test. A P -value less than 0.05 in this test indicates the presence of heterogeneity, suggesting variability in the causal effects estimated by different IVs [ 2 9] . E. Leave-One-Out Analysis: To ensure the robustness of our findings, we conducted a leave-one-out analysis. This involves sequentially excluding each SNP from the set of IVs and recalculating the causal estimate. This process helps identify any single SNP that might disproportionately influence the results, ensuring the reliability of our findings. All statistical analyses were conducted using R software (version 4.3.1), with specific analyses carried out using the "TwoSampleMR" and "forestploter" packages. These tools are specifically designed for MR analyses and provide a comprehensive suite of functions to perform various MR-related statistical tests. This thorough approach to statistical analysis ensures that our results are both reliable and valid, providing a strong foundation for our conclusions regarding the causal relationship between metformin treatment and AAA suppression. 2.3 Colocalization analysis study Colocalization refers to the spatial overlap of two or more entities within a biological sample, and it can be measured quantitatively using various statistical methods. Colocalization analysis is a statistical method used to test whether two input phenotypes are driven by the same genetic variant site in a certain region, thereby strengthening the evidence of association between the two phenotypes. These phenotypes can be molecular phenotypes, continuous traits or binary traits. Colocalization analysis can identify genes or gene variants that may play a role in multiple traits or diseases, thereby providing new insights into biological mechanisms or providing new targets for disease prevention and treatment. Colocalization analysis relies on four assumptions. For a certain genome interval, the assumptions are as follows: H0: Phenotype 1 and phenotype 2 are not significantly related to all SNP sites in a certain genomic region; H1/H2: Phenotype 1/phenotype 2 is significantly related to the SNP site in a certain genomic region, but phenotype 2/phenotype 1 has nothing to do with it; H3: Phenotype 1 and Phenotype 2 are significantly associated with SNP sites in a certain genomic region, but are driven by two independent causal variant sites; H4: Phenotype 1 and Phenotype 2 are significantly related to SNP sites in a certain genomic region and are driven by the same causal variant site. 3. Results 3.1 Study Inclusion for Meta-analysis Our study involved a comprehensive review and meta-analysis of the existing literature related to our research question. This process was characterized by an exhaustive search, rigorous screening, and the eventual inclusion of high-quality studies, as illustrated in Figure 2 . The literature search yielded 54 relevant articles in total. To ensure the validity and applicability of our study, we set specific inclusion and exclusion criteria for the articles. These criteria were meticulously designed based on the scope of our research, the quality of the studies, and the relevance of their findings to our research question. Each of the 54 articles underwent a thorough screening process. This involved examining the articles' titles and abstracts, reading the full texts where necessary, and evaluating their methodology and findings against our inclusion and exclusion criteria. This rigorous scrutiny ensured that only the most pertinent and high-quality studies were included in our review. After this rigorous screening process, a total of five articles(10-14) were deemed suitable for inclusion in our review. These five articles, in turn, comprised a total of 8 cohort studies, involving an impressive total of 14710 participants. The scale of these studies enhances the robustness and generalizability of our review's findings. The characteristics of each included study, such as the study design, sample size, population characteristics, exposure and outcome measures, and key findings, are presented in Table 2 . This table provides a snapshot of the breadth and depth of the studies included in our review, offering insights into the diverse methodologies and findings from which our conclusions are drawn. Importantly, the quality of the included studies was assessed using the Newcastle-Ottawa Scale (NOS), a widely recognized tool for evaluating the quality of non-randomized studies in meta-analyses. All seven articles scored greater than 7 on the NOS scale, indicating high quality. 3.2 Meta-analysis for association between metformin and enlargement of abdominal aortic diameter Firstly, this substantial sample size was split into two groups: one group of 598 participants had been prescribed metformin, while the other group of 8730 participants had not been prescribed metformin. Secondly, this information was used to categories patients into two groups: patients with no history or medically recorded diagnosis of diabetes (n = 1820) and patients previously diagnosed with diabetes prescribed metformin at the time of recruitment (n = 238). In Figure 3 A , our meta-analysis performed between diabetes prescribed metformin compared to diabetes not prescribed metformin and the progression of abdominal aortic diameter enlargement. Our forest plots, a graphical display designed to illustrate the relative strength of treatment effects in multiple studies, showed that metformin usage was associated with an inhibition of abdominal aortic diameter enlargement when compared to patients not treated with metformin. This was shown through a Mean Difference (MD) of -0.60 (95%CI: -0.97 - -0.23, P = 0.002), indicating a significantly lesser enlargement in the group treated with metformin. Although there is no publication bias ( Figure 3 B ), this high level of heterogeneity (I 2 statistic of 80% and P<0.0001) suggests that the studies varied considerably in their outcomes, which may reflect differences in study design, population characteristics, or other factors. In response to the high heterogeneity, we conducted a sensitivity analysis to identify potential sources of these issues. Following the exclusion of one cohort of the study by Golledge et al. (2019), patients with initial AAA diameter≤50 mm, the heterogeneity did not decrease (I 2 =83%, P<0.00001), as shown in Figure 2 . This was shown through MD=-0.61 (95%CI: -1.05 - -0.16, P = 0.008, Figure 3 C ), indicating a significantly lesser enlargement in the group treated with metformin, also there is no publication bias ( Figure 3 D ). Finally, our study compared diabetes prescribed metformin patients with no history or medically recorded diagnosis of diabetes patients, showed that metformin usage was associated with an inhibition of abdominal aortic diameter enlargement (MD=-1.00, 95%CI: -1.33 - -0.68, P < 0.000001, Figure 3 E ). Inversely, this meta-analysis, there is no significant heterogeneity was observed (I 2 = 24%, P = 0.27) as well as publication bias ( Figure 3 F ). 3.3 Causal Effects of Metformin Treatment on AAA The results of the Mendelian randomization (MR) analysis, summarized in Supplementary Table 8 , showed that the slope of the solid line in the scatter plot of genetic variation on metformin treatment and AAA incidence ( Figure 4 A ) indicated a negative correlation between metformin treatment and AAA. Single SNP analysis also supported this result overall ( Figure 4 D ). All five analyses yielded negative beta values. The IVW method demonstrated a negative association between metformin treatment and AAA (OR = 0.0108, 95%CI = 0.000204 to 0.572, P = 0.0254). The weighted mode analysis also produced consistent results (OR = 0.0024, 95% CI = 0.000008 to 0.735, P = 0.0451). However, it is important to note that both the weighted mean and simple mode methods did not yield significant positive results. 3.4 Causal Effects of Metformin Drug Target Genes on AAA To further investigate the potential causal effects of metformin-related genes on AAA, we conducted MR analyses on 40 genes associated with the 5 target genes of metformin. Using a significance threshold of P -value < 0.05, and taking the IVW method as the primary method, we identified evidence suggesting a decreased risk of AAA associated with increased expression of MCI-related genes NDUFAF1, NDUFAF3, NDUFB5, NDUFB9, NDUFB10, NDUFS5, and NDUFV2, AMPK-related gene PRKAG3, and MG3-related gene GPD2. Among them, the expression of NDUFAF1, NDUFAF3, NDUFB9, NDUFB10, NDUFV2, PRKAG3 and GPD2 is negatively correlated with the prevalence of AAA. On the other hand, the expression of NDUFB5 and NDUFS5 is positively correlated with the prevalence of AAA. A summary of the analysis results for all positive genes can be found in Supplementary Table 9 . 3.5 Sensitivity Analysis We conducted multiple sensitivity analyses to assess the stability and reliability of our findings. Cochran's Q test, used to evaluate heterogeneity, revealed some heterogeneity between metformin treatment and AAA SNPs (Q = 73.623, P = 0.018, Supplementary Table 8B ), while no heterogeneity was found between target gene SNPs and AAA. The symmetry of the funnel plot also indicated consistent results ( Figure 4 B ). MR-Egger regression test, used to test for horizontal pleiotropy, showed no evidence of horizontal pleiotropy (MR-Egger intercept = 0.0092; SE = 0.016; P = 0.564). Additionally, it is worth noting that leave-one-out sensitivity testing suggested a strong influence of SNP rs34872471 on the causal effect of metformin treatment on AAA ( Figure 4 C ). Overall, our conclusions regarding the causal relationship between metformin treatment and AAA are stable and reliable ( Figure 5 ). 3.6 Mediating effect analysis of serum lipids index Based on a two-step mediation analysis, we evaluated the potential mediating effects of HDL, LDL, TC and TG in the previously determined two-sample MR stages. The initial test confirmed the causal relationship between metformin treatment and TC, LDL, and between TC, LDL and AAA, but there was no significant causal association between metformin treatment and HDL, TG. This made it possible to study the mediating effects of TC and LDL. Subsequently, we assessed the mediation effects of TC and LDL. All mediation Mendelian randomization analysis used IVW as the primary method, the results are revealed in Table 3 . For LDL, the mediation effect ratio (IE_div_TE) was 28.84% with an OR of 0.271 (95% CI = 0.133-0.552); for TC, IE_div_TE was 25.17% with an OR of 0.3199 (95% CI = 0.159-0.644), indicating that there was a mediation effect between metformin and AAA mediated by LDL or TC. Supplementary Table 10 provides full results of the two-step analysis. 3.7 Colocalization analysis results of all nine positive genes Taking the P -value>0.75 of PP.H4 as the positive SNP site as the benchmark, the graphs drawn from the analysis results of all positive genes are shown in Figure 6 . For the nine genes with positive results in the Mendelian randomization analysis (GPD2, NDUFAF1, NDUFAF3, NDUFB5, NDUFB9, NDUFB10, NDUFS5, NDUFV2 and PRKAG3), the SNPs with the strongest causal associations are rs2711744, rs62019915, rs13010713, rs79975467, rs10745332, rs11761528, rs77992778, rs1442685 and rs1365963. Among these nine genes, GPD2 has significantly more positive SNPs, which indicates that GPD2 may be the most important site of action of metformin in inhibiting the onset of AAA. In addition, it is worth noting that the analysis results of NDUFS5 and NDUFV2 showed almost no SNPs that met the screening criteria. 4. Discussion Abdominal aortic aneurysm (AAA) remains a significant silent player in the pathogenesis of cardiovascular diseases, often going undiagnosed until rupture has occurred, leading to high mortality rates. Currently, the therapeutic interventions for AAA are majorly restricted to surgical approaches like open repair and endovascular aneurysm repair [ 3 , 33 , 34 ] . These techniques mainly focus on the symptomatic stages of AAA when the aneurysm has enlarged beyond a particular threshold. What is more concerning is that these surgical procedures accompany their fair share of complications such as graft infections, endoleaks, and renal impairment making the management of AAA challenging to clinicians [ 35 ] . Despite the rigorous research in the field of therapeutic methods, there is much left undiscovered and unexplored, especially when considering pharmaceutical treatments for AAA. Additionally, several trials have been performed to repurpose drugs like doxycycline and roxithromycin, suggesting potential benefits in AAA. Still, these results are yet to translate clinically due to insufficient evidence [ 36 ] . Furthermore, the use of other drugs, like losartan, has revealed disappointing results with no significant impact on AAA progression. Our exploration into the realm of metformin’s potential effect on AAA has led us through a broad range of methodologies, amplified our understanding of the connection between these two diverse yet integrally connected areas, and opened gateways to new avenues of research. The potential transformative implication of our study goes beyond the traditional clinical context and has significant echos in the research and policy arenas. The experimental evidence obtained through rodent models, though useful in deciphering mechanistic insights, often falls short in their translatability to humans due to differences in physiology, anatomy and disease mechanisms [ 37 , 38 ] . This gap provides an impetus for researchers to devise human-centered, clinically relevant research models that better mimic human disease paths. The challenges in uncovering genetic complexities, disease pathways, and treatment effects emphasize the need for the development of novel tools and approaches in predictive modeling. Our meta-analysis exhibits varying findings on the association between metformin and AAA. Some studies imply a protective effect of metformin against AAA, others provide inconclusive or even contrasting results [ 39 ] . These discrepancies underline the multifactorial nature of AAA, influenced not only by metformin but also by a host of confounding factors such as medical history, lifestyle, and genetic predisposition. Therefore, although our results suggest a potential correlation, they call for a cautious interpretation and advocate for a need for more rigorous to establish a precise causal relationship conclusively. It's worth noting that the Mendelian Randomization comes with its own set of assumptions, which if violated, can lead to biased results. Therefore, it’s crucial to conduct sensitivity analysis and use various MR methods to improve the robustness of our findings. Our study represents a cautious but forward step in this direction. There is an essential need for more research and replication of our findings in a wider range of populations with varying genetic backgrounds and lifestyle factors, to improve the generalizability of our conclusions [ 40 ] . Mechanically, our Mendelian randomization analysis has provided significant mechanistic insights into the potential action of metformin on AAA. AAA embodies a complex etiology involving matrix degradation, inflammation, and oxidative stress, making the development of an optimal pharmaceutical intervention challenging [ 41 , 42 ] . A promising avenue for cultivating novel AAA treatments involves gaining a deeper understanding of the specific genetic and molecular insurgents involved in AAA development and progression. Such insights can guide the research of new drugs targeting these specific mechanisms, paving the way for personalized medicine in AAA management. It appears that metformin potentially impacts AAA through several target genes such as AMPK, GDF15, MG53, FBP1, and MCI. AMPK (AMP-activated protein kinase), activated by metformin, is known for its role in energy metabolism and appears to be protective against AAA by triggering vasodilation, reducing vascular inflammation, and inhibiting vascular smooth muscle cell proliferation [ 43 , 44 ] . GDF15 (Growth Differentiation Factor 15) and MG53 (Mitsugumin 53), both upregulated by metformin, contribute to the control of metabolic dysfunction, inflammation, and tissue repair [ 45 , 46 ] . These processes are key components in the development and progression of AAA, thus highlighting their potential role in AAA prevention. FBP1 (Fructose-1,6-Bisphosphatase 1), a gene also targeted by metformin, is one of the essential regulatory enzymes in gluconeogenesis. Its regulation via metformin could have cardio-protective effects by increasing glucose uptake and reducing glucose production. Repurposing FBP1: dephosphorylating IκBα to suppress NF-κB [ 47 ] . Activation of NF-κB within endothelial cells leads to an increase in the expression of adhesion molecules. This can initiate the infiltration of macrophages and subsequent inflammation within the adventitia and media layers in AAA. Therefore, the endothelium is significantly involved in vascular remodeling and the development of aneurysms via its internal NF-κB signaling pathways [ 48 ] . Finally, MCI (Metabolic Complication Indicator), targeted by metformin, plays a role in signaling metabolic abnormalities, highlighting a possible role for metformin in mitigating AAA through metabolic regulation. In addition, according to further two-step mediation analysis, the key way for metformin to reduce the risk of AAA by regulating blood lipid levels should be to protect vascular smooth muscle cells from cholesterol-induced functional changes by lowering cholesterol levels, especially low-density lipoprotein cholesterol levels [ 49 ] . The pathway by which metformin regulates cholesterol levels is AMPK/SIRT1 [ 50 , 51 ] . However, despite these compelling mechanistic insights, more definitive evidence is needed to substantiate these associations. Hence, further research should focus on validating these target effects in experimental models and clinical trials, thereby providing a more in-depth understanding of the mechanisms underlying these associations. This approach would contribute significantly to the development of potential therapeutics aiming to leverage metformin’s impact on AAA. Strengths and limitations that come to light through the course of our research. Starting with strengths, our Mendelian randomization approach allows us to bypass some of the inherent challenges in conducting RCTs. This gives our study a significant advantage as it offers a relatively unbiased estimation of the causal effect - a critical stepping stone in the therapeutic investigation. Using genetic variants as instrumental variables allows us to estimate the effects of metformin therapy on AAA, providing us substantial evidence which otherwise would have been challenging to obtain. By blending an extensive literature search and synthesis, this approach ensures comprehensive coverage of existing evidence, minimizing the possibility of missing relevant information, and providing a thorough understanding of the research landscape surrounding metformin and AAA. The regulation of serum lipids can also be considered as a potential mechanism of action for the genetic causal association between metformin and AAA. However, there are also limitations to be acknowledged. While MR studies mimic the randomizing nature of RCTs, they are not entirely free of biases. It's noteworthy that MR relies heavily on the assumptions it makes about the genetic variants used. Any violation of these assumptions can introduce biases that may over or underestimate the true effect. Furthermore, despite the comprehensive literature search in the meta-analysis phase, the possibility of oversight of unpublished studies and articles not indexed in the database may introduce publication bias. Lastly, while we have pinpointed potential targets, the biological pathways involving metformin and AAA are highly complex and our understanding is still at a nascent stage. Our study proposes a pathway of investigation and does not establish a definitive cause-and-effect relationship. 5. Conclusion This study provides compelling evidence that metformin, traditionally used for type II diabetes, holds promise as a therapeutic agent for Abdominal Aortic Aneurysm (AAA). Our findings, derived from meta-analysis, Mendelian randomization, and colocalization analysis, suggest that metformin can significantly inhibit AAA progression. Notably, colocalization analysis identified specific SNPs in genes such as GPD2 that are associated with both metformin targets and AAA, underscoring a genetic basis for this therapeutic effect. Furthermore, the mediating effect of metformin on lipid profiles, particularly its ability to lower LDL and total cholesterol levels, highlights an important pathway through which metformin may exert its protective effects. These insights advocate for further clinical trials to validate metformin's efficacy in AAA treatment, potentially offering a non-surgical option for managing this life-threatening condition. Declarations Acknowledgments : This work was supported by the Fundamental Research Funds for the Central Universities (grant number: DUT22YG107), the National Natural Science Foundation of China (grant number: 81600370), the China Postdoctoral Science Foundation (grant number: 2018M640270) and the Natural Science Foundation of Liaoning Province (2023-MS-096) for Yanshuo Han, also supported by National Natural Science Foundation of China (grant number: 81970402 and 82170507) for Jian Zhang, and supported by International Science and Technology Cooperation Program of Liaoning Province (grant number: 2023JH2/10700019). We would like to acknowledge all the above institutions that provided financial support. In addition, We would like to acknowledge the patients and medical institutions who selflessly contributed valuable publicly available data to this study. 6. Declaration of financial/other relationships The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. 7. Author contributions Z.Z. found the data, performed the MR analysis and interpreted all sections regarding the methods and results of the MR analysis. Y.H., J.U. and A.H. designed the study and drafted the manuscript. J.W., X.Z., D.J., Y.L., H.J. and J.Z. contributed other parts including bibliometric and meta-analysis, and checked and revised the manuscript. 8. Ethical approval Each cohort included in this study has been conducted using published studies and consortia providing publicly available summary statistics, so no additional ethical approval was required.. All original studies has received ethical approval and agreed to participate, and summary-level data were provided for analysis. 9. Data availability statement All of the data generated or analyzed in this study are either contained in the article/supplementary material or in the data repositories listed in References. Further inquiries can be directed to the corresponding authors. References Ahmed Z, Zeeshan S, Mendhe D, Dong X. Human gene and disease associations for clinical-genomics and precision medicine research. Clin Transl Med. 2020;10(1):297-318. Baman JR, Eskandari MK. What Is an Abdominal Aortic Aneurysm? JAMA. 2022;328(22):2280. Han Y, Zhang S, Zhang J, Ji C, Eckstein HH. Outcomes of Endovascular Abdominal Aortic Aneurysm Repair in Octogenarians: Meta-analysis and Systematic Review. Eur J Vasc Endovasc Surg. 2017;54(4):454-63. Siordia JA. Beta-Blockers and Abdominal Aortic Aneurysm Growth: A Systematic Review and Meta-Analysis. Curr Cardiol Rev. 2021;17(4):e230421187502. Hosseini A, Sahranavard T, Reiner Z, Jamialahmadi T, Dhaheri YA, Eid AH, et al. Effect of statins on abdominal aortic aneurysm. Eur J Pharm Sci. 2022;178:106284. Thatcher SE, Zhang X, Howatt DA, Yiannikouris F, Gurley SB, Ennis T, et al. Angiotensin-converting enzyme 2 decreases formation and severity of angiotensin II-induced abdominal aortic aneurysms. Arterioscler Thromb Vasc Biol. 2014;34(12):2617-23. Goldberg RB, Orchard TJ, Crandall JP, Boyko EJ, Budoff M, Dabelea D, et al. Effects of Long-term Metformin and Lifestyle Interventions on Cardiovascular Events in the Diabetes Prevention Program and Its Outcome Study. Circulation. 2022;145(22):1632-41. Li JZ, Li YR. Cardiovascular Protection by Metformin: Latest Advances in Basic and Clinical Research. Cardiology. 2023;148(4):374-84. Tai S, Sun J, Zhou Y, Zhu Z, He Y, Chen M, et al. Metformin suppresses vascular smooth muscle cell senescence by promoting autophagic flux. J Adv Res. 2022;41:205-18. Fujimura N, Xiong J, Kettler EB, Xuan H, Glover KJ, Mell MW, et al. Metformin treatment status and abdominal aortic aneurysm disease progression. J Vasc Surg. 2016;64(1):46-54 e8. Golledge J, Morris DR, Pinchbeck J, Rowbotham S, Jenkins J, Bourke M, et al. Editor's Choice - Metformin Prescription is Associated with a Reduction in the Combined Incidence of Surgical Repair and Rupture Related Mortality in Patients with Abdominal Aortic Aneurysm. Eur J Vasc Endovasc Surg. 2019;57(1):94-101. Golledge J, Moxon J, Pinchbeck J, Anderson G, Rowbotham S, Jenkins J, et al. Association between metformin prescription and growth rates of abdominal aortic aneurysms. Br J Surg. 2017;104(11):1486-93. Itoga NK, Rothenberg KA, Suarez P, Ho TV, Mell MW, Xu B, et al. Metformin prescription status and abdominal aortic aneurysm disease progression in the U.S. veteran population. J Vasc Surg. 2019;69(3):710-6 e3. Unosson J, Wagsater D, Bjarnegard N, De Basso R, Welander M, Mani K, et al. Metformin Prescription Associated with Reduced Abdominal Aortic Aneurysm Growth Rate and Reduced Chemokine Expression in a Swedish Cohort. Ann Vasc Surg. 2021;70:425-33. Hariton E, Locascio JJ. Randomised controlled trials - the gold standard for effectiveness research: Study design: randomised controlled trials. BJOG. 2018;125(13):1716. Schooling CM, Freeman G, Cowling BJ. Mendelian randomization and estimation of treatment efficacy for chronic diseases. Am J Epidemiol. 2013;177(10):1128-33. Sanderson E, Glymour MM, Holmes MV, Kang H, Morrison J, Munafo MR, et al. Mendelian randomization. Nat Rev Methods Primers. 2022;2. Davey Smith G, Hemani G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet. 2014;23(R1):R89-98. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol. 2010;25(9):603-5. DerSimonian R, Laird N. Meta-analysis in clinical trials revisited. Contemp Clin Trials. 2015;45(Pt A):139-45. Cordero CP, Dans AL. Key concepts in clinical epidemiology: detecting and dealing with heterogeneity in meta-analyses. J Clin Epidemiol. 2021;130:149-51. Kriston L. Dealing with clinical heterogeneity in meta-analysis. Assumptions, methods, interpretation. Int J Methods Psychiatr Res. 2013;22(1):1-15. Zheng J, Xu M, Walker V, Yuan J, Korologou-Linden R, Robinson J, et al. Evaluating the efficacy and mechanism of metformin targets on reducing Alzheimer’s disease risk in the general population: a Mendelian randomisation study. Diabetologia. 2022;65(10):1664-75. Zheng J, Xu M, Yang Q, Hu C, Walker V, Lu J, et al. Efficacy of metformin targets on cardiometabolic health in the general population and non-diabetic individuals: a Mendelian randomization study. EBioMedicine. 2023;96:104803. Carter AR, Gill D, Davies NM, et al. Understanding the consequences of education inequality on cardiovascular disease: mendelian randomisation study. BMJ. 2019;365:l1855. Published 2019 May 22. doi:10.1136/bmj.l1855 Burgess S, Thompson SG. Avoiding bias from weak instruments in Mendelian randomization studies. Int J Epidemiol. 2011;40(3):755–64. Burgess S, Scott RA, Timpson NJ, Davey Smith G, Thompson SG, Consortium E-I. Using published data in Mendelian randomization: a blueprint for efficient identification of causal risk factors. Eur J Epidemiol. 2015;30(7):543-52. Burgess S, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol. 2013;37(7):658-65. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol. 2015;44(2):512-25. Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median Estimator. Genet Epidemiol. 2016;40(4):304-14. Hartwig FP, Davey Smith G, Bowden J. Robust inference in summary data Mendelian randomization via the zero modal pleiotropy assumption. Int J Epidemiol. 2017;46(6):1985-98. Soo Hoo AJ, Fitzgibbon JJ, Hussain MA, Scully RE, Servais AB, Nguyen LL, et al. Contemporary indications for open abdominal aortic aneurysm repair in the endovascular era. J Vasc Surg. 2022;76(4):923-31 e1. Deery SE, Schermerhorn ML. Open versus endovascular abdominal aortic aneurysm repair in Medicare beneficiaries. Surgery. 2017;162(4):721-31. Niaz OS, Rao A, Abidia A, Parrott R, Refson J, Somaiya P. Surgical and medical interventions for abdominal aortic graft infections. Cochrane Database Syst Rev. 2020;8(8):CD013469. Sterpetti AV. Doxycycline and Growth of Abdominal Aortic Aneurysms. JAMA. 2020;324(15):1568-9. Wang Z, Guo J, Han X, Xue M, Wang W, Mi L, et al. Metformin represses the pathophysiology of AAA by suppressing the activation of PI3K/AKT/mTOR/autophagy pathway in ApoE(-/-) mice. Cell Biosci. 2019;9:68. Guo J, Wang Z, Xue M, Mi L, Zhao M, Ma C, et al. Metformin protects against abdominal aortic aneurysm by Atg7-induced autophagy. Adv Clin Exp Med. 2022;31(1):59-69. Yuan Z, Heng Z, Lu Y, Wei J, Cai Z. The Protective Effect of Metformin on Abdominal Aortic Aneurysm: A Systematic Review and Meta-Analysis. Front Endocrinol (Lausanne). 2021;12:721213. Liu H, Wu W, Xiang W, Yuan J. Lifestyle factors, metabolic factors and socioeconomic status for pelvic organ prolapse: a Mendelian randomization study. Eur J Med Res. 2023;28(1):183. Wu Y, Jiang D, Zhang H, Yin F, Guo P, Zhang X, et al. N1-Methyladenosine (m1A) Regulation Associated With the Pathogenesis of Abdominal Aortic Aneurysm Through YTHDF3 Modulating Macrophage Polarization. Front Cardiovasc Med. 2022;9:883155. Stepien KL, Bajdak-Rusinek K, Fus-Kujawa A, Kuczmik W, Gawron K. Role of Extracellular Matrix and Inflammation in Abdominal Aortic Aneurysm. Int J Mol Sci. 2022;23(19). He J, Li N, Fan Y, Zhao X, Liu C, Hu X. Metformin Inhibits Abdominal Aortic Aneurysm Formation through the Activation of the AMPK/mTOR Signaling Pathway. J Vasc Res. 2021;58(3):148-58. Wang S, Zhang C, Zhang M, Liang B, Zhu H, Lee J, et al. Activation of AMP-activated protein kinase alpha2 by nicotine instigates formation of abdominal aortic aneurysms in mice in vivo. Nat Med. 2012;18(6):902-10. Sanchez-Infantes D, Nus M, Navas-Madronal M, Fite J, Perez B, Barros-Membrilla AJ, et al. Oxidative Stress and Inflammatory Markers in Abdominal Aortic Aneurysm. Antioxidants (Basel). 2021;10(4). Cui KM, Hu ZP, Wang YL. MG53 represses high glucose-induced inflammation and angiogenesis in human retinal endothelial cells by repressing the EGR1/STAT3 axis. Immunopharmacol Immunotoxicol. 2022;44(4):484-91. Zhang G, Tao J, Lin L, Qiu W, Lu Z. Repurposing FBP1: dephosphorylating IkappaBalpha to suppress NFkappaB. Cell Res. 2023;33(6):419-20. Saito T, Hasegawa Y, Ishigaki Y, Yamada T, Gao J, Imai J, et al. Importance of endothelial NF-kappaB signalling in vascular remodelling and aortic aneurysm formation. Cardiovasc Res. 2013;97(1):106-14.February 2006. Solymár M, Ivic I, Pótó L, et al. Metformin induces significant reduction of body weight, total cholesterol and LDL levels in the elderly - A meta-analysis. PLoS One. 2018;13(11):e0207947. Published 2018 Nov 26. doi:10.1371/journal.pone.0207947 Xing H, Liang C, Wang C, Xu X, Hu Y, Qiu B. Metformin mitigates cholesterol accumulation via the AMPK/SIRT1 pathway to protect osteoarthritis chondrocytes. Biochem Biophys Res Commun. 2022;632:113-121. doi:10.1016/j.bbrc.2022.09.074 Ali A, Unnikannan H, Shafarin J, et al. Metformin enhances LDL-cholesterol uptake by suppressing the expression of the pro-protein convertase subtilisin/kexin type 9 (PCSK9) in liver cells. Endocrine. 2022;76(3):543-557. doi:10.1007/s12020-022-03022-x Tables Table 1 Source of GWAS data for metformin treatment and AAA and eQTL data for five metformin targets Phenotype Year Author Samplesize or case/control Ancestry Web Source if Publicly Available Exposure Metformin treatment 2018 Ben Elsworth 11,552/451,381 European https://gwas.mrcieu.ac.uk/datasets/ukb-b-14609 Target Genes 2018 Urmo Vosa 31,684 European https://gwas.mrcieu.ac.uk/datasets/eqtl-a-(Corresponding ENSG ID), Please refer to: https://useast.ensembl.org/index.html Mediators LDL-C 2022 Tom Richardson 115,082 European https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST90092883/ HDL-C 2022 Tom Richardson 115,082 European https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST90092822/ TG 2022 Tom Richardson 115,082 European https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST90092992/ TC 2022 Tom Richardson 115,082 European https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST90092985/ Outcome Abdominal aortic aneurysm 2021 NA 4,217/469,464 European https://risteys.finregistry.fi/endpoints/I9_ABAORTANEUR Table 2 Summary of Included Studies Authors Year Country Study design Cases Participants Drug Follow-up, Years Outcomes NOS Fujimura Naoki 2016 United States Cohort study 58 Elderly patients with diabetes with untreated AAA 1. Metformin (T2DM); 2. Without metformin (T2DM) 2.6 Maximum aortic diameter growth measured by contrast-enhanced CT 7 Golledge Jonathan 2019 Australia Cohort study A 1,080 Patients with asymptomatic unrepaired AAA of any diameter≥30mm 1. Diabetes with no history or medically recorded； 2. Diabetes prescribed metformin at the time of recruitment; 3. Diabetes not prescribed metformin. 3.2 The combined incidence of AAA repair or mortality due to AAA rupture 8 Golledge Jonathan 2019 Australia Cohort study B 763 Patients with asymptomatic unrepaired AAA of any diameter ≥ 50mm 1. Diabetes with no history or medically recorded； 2. Diabetes prescribed metformin at the time of recruitment; 3. Diabetes not prescribed metformin at the time of recruitment. 3.6 The combined incidence of AAA repair or mortality due to AAA rupture 8 Golledge Jonathan 2017 Australia and New Zealand Cohort study A 1357 Elderly patients with AAA (infrarenal aortic diameter≥ 30mm) 1. Diabetes who were prescribed metformin； 2. Diabetes not prescribed metformin; 3. Patients who neither had diabetes nor were receiving metformin 3.6 Maximum aortic diameter growth measured by contrast enhanced CT 8 Golledge Jonathan 2017 Australia and New Zealand Cohort study B 287 Elderly patients with AAA (infrarenal aortic diameter≥ 30mm) 1. Diabetes who were prescribed metformin; 2. Diabetes not prescribed metformin; 3. Patients who neither had diabetes nor were receiving metformin 2.9 Maximum aortic diameter growth measured by contrast enhanced CT 8 Golledge Jonathan 2017 Australia and New Zealand Cohort study C 53 Elderly patients with AAA (infrarenal aortic diameter≥ 30mm) 1. Diabetes who were prescribed metformin; 2. Diabetes not prescribed metformin; 3. Patients who neither had diabetes nor were receiving metformin 1 Maximum aortic diameter growth measured by contrast enhanced CT 8 Itoga Nathan K 2019 United States Cohort study 13,843 Elderly patients with diabetes and a diagnosis of AAA without rupture 1. Diabetes prescribed metformin 2. Diabetes not prescribed metformin 4.2 Maximum aortic diameters growth determined from radiographic reports 8 Unosson Jon 2021 Sewden Cohort study 98 Patients with initial abdominal aortic diameter ≥30mm 1. Metformin (T2DM) 2. Without metformin (T2DM) 3.2 Maximum aortic diameter growth measured by ultrasound 8 Table 3 Mediating effects of TC and LDL Mediator OR of Intermediary effect (95% Cl) b_mediation se_mediation Mediation efect proportion LDL 0.271 (0.133-0.552) -1.3057 0.3627 0.2884 TC 0.3199 (0.159-0.644) -1.1397 0.3572 0.2517 Additional Declarations No competing interests reported. Supplementary Files SupplementaryTable1.docx SupplementaryTable2.docx SupplementaryTable3.docx SupplementaryTable4.docx SupplementaryTable5.docx SupplementaryTable6.docx SupplementaryTable7.docx SupplementaryTable8.docx SupplementaryTable9.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6834737","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":476935149,"identity":"299bf5bd-b812-4e6a-9e27-86e2f48fb665","order_by":0,"name":"Zhaoxuan Zhang","email":"","orcid":"","institution":"Dalian University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Zhaoxuan","middleName":"","lastName":"Zhang","suffix":""},{"id":476935150,"identity":"1ee5c15d-8be9-4048-98d5-f4eca03b7cef","order_by":1,"name":"Yuemeng Li","email":"","orcid":"","institution":"Central Hospital of Dalian University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Yuemeng","middleName":"","lastName":"Li","suffix":""},{"id":476935151,"identity":"221e784a-8391-4887-bc5f-80cb3073309e","order_by":2,"name":"Jian Wang","email":"","orcid":"","institution":"Dalian University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Jian","middleName":"","lastName":"Wang","suffix":""},{"id":476935152,"identity":"61517c07-cc7d-460d-b1b0-e6eb132d3f54","order_by":3,"name":"Xiaoxu Zhang","email":"","orcid":"","institution":"Dalian University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Xiaoxu","middleName":"","lastName":"Zhang","suffix":""},{"id":476935153,"identity":"1d239ed1-9f36-479a-b4c7-524852ca2a27","order_by":4,"name":"Deying Jiang","email":"","orcid":"","institution":"Central Hospital of Dalian University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Deying","middleName":"","lastName":"Jiang","suffix":""},{"id":476935154,"identity":"482998a3-aa44-4c7a-a141-27da78e809ab","order_by":5,"name":"Azad Hussain","email":"","orcid":"","institution":"University of Gujrat","correspondingAuthor":false,"prefix":"","firstName":"Azad","middleName":"","lastName":"Hussain","suffix":""},{"id":476935155,"identity":"9c29637b-778b-4c1a-9e3d-dc199409db11","order_by":6,"name":"Jamol Uzokov","email":"","orcid":"","institution":"Republican Specialized Scientific Practical Medical Center of Therapy and Medical Rehabilitation","correspondingAuthor":false,"prefix":"","firstName":"Jamol","middleName":"","lastName":"Uzokov","suffix":""},{"id":476935156,"identity":"e92a67e8-93e0-4fdc-bf2d-abefa31ac554","order_by":7,"name":"Han Jiang","email":"","orcid":"","institution":"First Hospital of China Medical University","correspondingAuthor":false,"prefix":"","firstName":"Han","middleName":"","lastName":"Jiang","suffix":""},{"id":476935157,"identity":"519ab24a-3b0b-46dd-b5d3-30665ae707fb","order_by":8,"name":"Jian Zhang","email":"","orcid":"","institution":"First Hospital of China Medical University","correspondingAuthor":false,"prefix":"","firstName":"Jian","middleName":"","lastName":"Zhang","suffix":""},{"id":476935158,"identity":"f981b6a5-c80d-4c49-b73d-645e329078b1","order_by":9,"name":"Yanshuo Han","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6ElEQVRIiWNgGAWjYJACZiBOAFIHJB6A+QlEa2FLkEhgMCBJC48BcVoMjp89/LmwzSaPf3bPxxuJbX8Y+NlzDBh+7sCj5UxemvTMtrRiiTtnN1skthkwSPa8MWDsPYNHy4EcM2betsOJDTdyt0mAtBjcyDFgZmzDo+X8G+PPIC3zb+Q8A2uxJ6gFqEAapGXDjRw2iC0SBLRI3nhjJs1zLi1x4400Y4uEc8Y8EmeeFRzsxaOF73yO8WeeMpvEeTeSH974UCYnx9+evPHBTzxaFA6gCfCACHRBFCDfgE92FIyCUTAKRgEIAAA5JFQARn5fBwAAAABJRU5ErkJggg==","orcid":"","institution":"Dalian University of Technology","correspondingAuthor":true,"prefix":"","firstName":"Yanshuo","middleName":"","lastName":"Han","suffix":""}],"badges":[],"createdAt":"2025-06-06 07:38:36","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6834737/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6834737/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":85566073,"identity":"fa5e56f1-bc94-4536-86b8-194483c9c770","added_by":"auto","created_at":"2025-06-27 14:35:36","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":692958,"visible":true,"origin":"","legend":"\u003cp\u003eStudy design and workflow. GWAS, genome-wide association study; IVW, inverse variance weighted; MR, Mendelian randomization; AAA, abdominal aortic aneurysm.\u003c/p\u003e","description":"","filename":"Fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/5120f485a380c6857416c5c9.png"},{"id":85566074,"identity":"328e61fa-3520-48f4-902d-aca7a35db2e8","added_by":"auto","created_at":"2025-06-27 14:35:36","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":745634,"visible":true,"origin":"","legend":"\u003cp\u003ePreferred Reporting Items for Systematic reviews and Meta-Analyses diagram illustrating the selection of the included trial.\u003c/p\u003e","description":"","filename":"Fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/e9351fe4b3e509a8b3c939ba.png"},{"id":85566870,"identity":"8f591ea6-c249-43f8-8959-9fe26e265fc0","added_by":"auto","created_at":"2025-06-27 14:51:36","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":727804,"visible":true,"origin":"","legend":"\u003cp\u003eForest plots of studies comparing the effect of diabetes prescribed metformin versus diabetes non-metformin on progression of abdominal aortic diameter enlargement (A) and sensitivity analysis results (C); Forest plots of studies comparing the effect of diabetes prescribed metformin versus patients with no history or medically recorded diagnosis of diabetes on progression of abdominal aortic diameter enlargement (E), the diamond represents the aggregated effect estimate and its 95% CI; Funnel plots corresponding to the forest plots in (B), (D), and (F), respectively. These plots assess publication bias by plotting the treatment effect against a measure of study size. The presence of asymmetry can suggest potential publication bias.\u003c/p\u003e","description":"","filename":"Fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/75af0287d28e687c3512a682.png"},{"id":85566651,"identity":"3d036f51-6611-4c24-9029-04c01645f2a0","added_by":"auto","created_at":"2025-06-27 14:43:36","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":607988,"visible":true,"origin":"","legend":"\u003cp\u003eScatter plot of the effect of genetic variation on metformin treatment and AAA (A). The slope of the solid line represents the magnitude of the association estimated from the MR analysis; Funnel plot of the causal effect of metformin treatment on AAA (B); Fixed-effects IVW analysis of the causal effect of metformin on AAA. Black dots and bars represent causal estimates and 95%CI using each SNP. Red dots and bars represent the overall estimate and 95% CI of the meta-analysis by fixed-effects IVW method (C); Leave-one-out analysis plot of metformin in AAA. SNP, single nucleotide polymorphism; CI, confidence interval (D).\u003c/p\u003e","description":"","filename":"Fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/d0ded02c0e41e5f59f99ddbc.png"},{"id":85566659,"identity":"aea90658-d3e0-4910-af76-23c3a650808c","added_by":"auto","created_at":"2025-06-27 14:43:37","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":456023,"visible":true,"origin":"","legend":"\u003cp\u003eMendelian randomization forest plots estimating genetically predicted associations between metformin treatment and AAA. OR, odds ratio; CI, confidence interval; IVW, inverse variance weighted. Points represent odds ratios for each standard deviation increment in metformin treatment. Error bars represent 95% confidence intervals.\u003c/p\u003e","description":"","filename":"Fig5.png","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/9fde9f95c552aea894a8b5d6.png"},{"id":85566652,"identity":"2197a8c9-f999-4f3a-8e80-2e3cb53ecdbc","added_by":"auto","created_at":"2025-06-27 14:43:36","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":3957335,"visible":true,"origin":"","legend":"\u003cp\u003eVisualization of colocalization analysis results. The eQTLs from A to I are: NDUFAF1, NDUFAF3, NDUFB5, NDUFB9, NDUFB10, NDUFS5, NDUFV2, PRKAG3 and GPD2.\u003c/p\u003e","description":"","filename":"Fig6.png","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/b2f17aec66fc980b111bfeeb.png"},{"id":88060908,"identity":"8936cb2b-9241-43d1-a851-28f44ac44041","added_by":"auto","created_at":"2025-08-01 01:46:41","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":7668323,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/9fde697a-8b2d-45fc-be5a-803aaab1354e.pdf"},{"id":85566650,"identity":"0fa3e795-ba13-4de6-803d-043055a20dce","added_by":"auto","created_at":"2025-06-27 14:43:36","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":25479,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable1.docx","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/459536eeb4386b9acbd6a2f6.docx"},{"id":85566075,"identity":"f34b6c2d-dd68-42c4-9359-58520536f954","added_by":"auto","created_at":"2025-06-27 14:35:36","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":13902,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable2.docx","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/de935f98c516c9e0e2a99f7b.docx"},{"id":85566077,"identity":"84b63a69-1811-4ff5-8221-c8c80cba0866","added_by":"auto","created_at":"2025-06-27 14:35:36","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":29429,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable3.docx","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/5665b359b5a77a2ef98efa12.docx"},{"id":85566082,"identity":"4b54a634-5eef-4c60-8db7-894e96e27711","added_by":"auto","created_at":"2025-06-27 14:35:36","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":111659,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable4.docx","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/1f15c1afdc24a2c500d2978f.docx"},{"id":85566085,"identity":"d0c28a31-d033-4002-9c5e-a52d80208da1","added_by":"auto","created_at":"2025-06-27 14:35:36","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":13975,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable5.docx","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/2f4d8ac0c4fb8cf79eda4c1d.docx"},{"id":85566872,"identity":"36bc9bd5-384d-4eb6-b2fb-16aaf5b6464d","added_by":"auto","created_at":"2025-06-27 14:51:36","extension":"docx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":20605,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable6.docx","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/ddf6756f6d77d73baba61f5c.docx"},{"id":85566094,"identity":"5dd33cb3-c1b6-4c4d-8066-cca594a4b9b0","added_by":"auto","created_at":"2025-06-27 14:35:36","extension":"docx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":16038,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable7.docx","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/718c768cb6f4ff98831585f8.docx"},{"id":85566112,"identity":"249377bc-fd4a-449c-b131-bc639732f150","added_by":"auto","created_at":"2025-06-27 14:35:37","extension":"docx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":27223,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable8.docx","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/f53b32889aeef06dff12a41f.docx"},{"id":85566120,"identity":"0cf237fe-be67-40ad-bd69-8720befc53a3","added_by":"auto","created_at":"2025-06-27 14:35:37","extension":"docx","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":24544,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable9.docx","url":"https://assets-eu.researchsquare.com/files/rs-6834737/v1/033433cfe51137415d4acef4.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Deciphering the Role of Metformin in Abdominal Aortic Aneurysm Progression: Insights from Two-Step Mendelian Randomization and Colocalization Analysis","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eUnderstanding the multifaceted role Interactions of biological factors play in the development and progression of diseases is a fundamental aspect of biomedical research\u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]\u003c/sup\u003e. An aspect that remains underexplored is the impact and relationship between diabetes treatment, particularly metformin, and cardiovascular disease. A specific area of focus in cardiovascular disorders is Abdominal Aortic Aneurysm (AAA), a condition of multifactorial etiology characterized by a progressive enlargement of the aorta, leading to aortic rupture if left undetected or untreated\u003csup\u003e[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]\u003c/sup\u003e. Abdominal aortic aneurysm (AAA) is a grave life-threatening cardiovascular disorder characterized by abnormal dilation of the aorta. This rupture-prone condition often remains asymptomatic until there is a rupture, leading to internal bleeding and potential fatality. The lack of specific symptoms poses a significant challenge in timely diagnosis and effective disease management, signaling an unmet need for novel detection and treatment strategies. Currently, the therapeutic interventions for AAA are majorly restricted to surgical approaches like open repair and endovascular aneurysm repair\u003csup\u003e3\u003c/sup\u003e. The pharmaceutical landscape for AAA is unfortunately sparse. Some recommendations include beta-blockers\u003csup\u003e[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]\u003c/sup\u003e, statins\u003csup\u003e[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]\u003c/sup\u003e, antiplatelet agents, and angiotensin-converting enzyme (ACE) inhibitors\u003csup\u003e[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]\u003c/sup\u003e which primarily focus on managing the risk factors, such as hypertension and hyperlipidemia that contribute to AAA incidence. However, their efficacy in hindering AAA growth and rupture remains inconclusive, warranting more research.\u003c/p\u003e \u003cp\u003eMetformin, the first-line therapy for type II diabetes, has been pivotal in the control and management of the disease, contributing immensely to therapeutic regimens due to its commendable safety profile, and high efficacy. Still, further illumination on its impact in the arena of cardiovascular health, though well established, remains in the early exploration phases\u003csup\u003e[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]\u003c/sup\u003e. The promising protective roles of metformin in cardiovascular health open avenues and pique both clinical and academic interest in further determining its potential as a useful tool in fighting cardiovascular diseases\u003csup\u003e[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]\u003c/sup\u003e. Appreciating the medicine's impact beyond its primary role as an oral hypoglycemic agent, addressing insulin resistance, might yield novel therapeutic and prognostic pathways in other diseases. However, the literature concerning its effectiveness in AAA treatment has shown mixed results, opening gates for further investigation\u003csup\u003e[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/sup\u003e. A multitude of studies, conducted primarily in rodent models, have delved into the impact of metformin on AAA\u003csup\u003e[\u003cspan additionalcitationids=\"CR11 CR12 CR13\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/sup\u003e. While these studies have shed light onto the possible effects, they have their limitations. Firstly, rodent models, although scientifically valuable, do not perfectly mimic the human pathology in terms of anatomy, physiology, and disease mechanisms. Therefore, the translational potential of these findings to humans might be restricted, thus calling for caution in interpretation. Secondly, another barrier is the dearth of Randomized Controlled Trials (RCTs). RCTs, often considered the gold standard in clinical research, provide powerful and reliable evidence on the efficacy of interventions. Given their experimental nature, RCTs often yield high-quality outcomes and are less prone to bias\u003csup\u003e[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/sup\u003e. They allow the demonstration of the cause and effect, providing robust evidence on efficacy and effectiveness. However, there remains a dearth of such trials intending to investigate the therapeutic potential of drugs like metformin in AAA management. Various factors, including resource constraints and the complexity of disease mechanisms, contribute to the challenges inhibiting the execution of RCTs.\u003c/p\u003e \u003cp\u003eEnter Mendelian randomization (MR), an analytical method using genetic variants as instrumental variables\u003csup\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/sup\u003e. It has been proposed as a solution to circumnavigate the above-mentioned challenges. By utilizing genetic variants that are associated with an exposure (in our case, metformin treatment), MR studies can help estimate the causal effect of the exposure on the outcome (AAA in this context). By harnessing the principles of genetics, where the alleles of genetic variants are randomly assigned at conception, MR studies mimic the randomization process naturally occurring in RCTs\u003csup\u003e[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/sup\u003e. In other words, Mendelian randomization can estimate the causal effect of specific modifiable exposures on disease outcomes, providing evidence that can complement or, in some cases, substitute the need for performing RCTs\u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/sup\u003e. This methodology is particularly useful for investigating questions that may be challenging or impossible to address with RCTs due to ethical or practical reasons. Its application can, therefore, enable us to scrutinize the impact of metformin in AAA, thereby filling the gap left by the absence of RCTs.\u003c/p\u003e \u003cp\u003eOur study is nested in this research gap, intent on exploring the impact of metformin therapy on the progression and clinical events related to AAA, particularly in patients without diabetes. The objective is to dissect the relationship that exists between metformin and AAA, to establish a probable causative effect, if any. This step is crucial as it forms the background on which we can build our understanding of the magnitude of the impact and subsequently design interventions. Safety, quality of life, and overall survival of patients may hinge partly on this valuable knowledge. By embarking on our study with MR, we aim to delve deeper into the complex interplay between AAA and metformin, uncovering insights that may potentially revolutionize therapeutic approaches to AAA. The limitations and advantages, findings and interpretations, and implications for future research will be discussed in detail in the sections to follow.\u003c/p\u003e"},{"header":"2. Material and methods","content":"\u003cp\u003e\u003cstrong\u003e2.1 Meta-analysis for association between metformin and abdominal aortic aneurysm\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.1.1 Literature Search Strategy\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn accordance with the PRISMA guidelines, an exhaustive literature search was executed across Medline, Embase, Cochrane Library, and Ovid databases, covering a timeframe from March 1, 1999 to November 11, 2023\u003csup\u003e[19]\u003c/sup\u003e. The search was centered around the key term \"aortic aneurysm, abdominal\". To ensure a comprehensive coverage, the search was broadened to include an array of synonyms and related terms for Metformin, such as Dimethylbiguanidine, Dimethylguanylguanidine, Glucophage, Metformin Hydrochloride, and others. The search strategy also encompassed additional terms such as \"Aortic aneurysm (Aneurysms, Aortic, Aortic Aneurysms, Aneurysm, Aortic),\" \"Aortic dilatation,\" to ensure all relevant studies were captured. To eliminate any bias and ensure objectivity, two independent reviewers were engaged to screen every article and select cohort studies for inclusion. This comprehensive search was not limited by language, ensuring a global perspective in the selection of studies. The resources scrutinized for this analysis spanned published articles, conference abstracts accompanied by statistical data and charts, as well as pertinent literature cited in the reference lists of the included articles. In cases where additional data was deemed necessary for a more comprehensive analysis, authors were contacted directly. We ensured that every piece of relevant information was captured, thereby enhancing the robustness and reliability of our meta-analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.1.2 Eligibility Criteria\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe eligibility of the articles was determined by two independent reviewers. Any conflicts between the reviewers were resolved by consulting a third independent reviewer. The studies considered for inclusion were randomized controlled trials, cohort studies, and case-control studies that satisfied the following criteria: The studies focused on AAA patients who were on Metformin’s medication, and they reported the AAA growth rate. Conference abstracts were included if they provided sufficient data for analysis. We deemed studies as eligible if they were cohort studies involving adult diabetic patients who had undergone a minimum of 8 weeks of metformin pharmacologic intervention. We excluded (1) reviews, case reports, comments, recommendations, letters, ongoing trials, protocols, conference abstracts, consensus or statements, and articles that lacked applicable data; (2) duplicate reports and studies of low quality, inconsistent type, or those providing insufficient information; and (3) studies using inappropriate statistical methods or providing insufficient data. The selected articles needed to clearly demonstrate the effect of metformin pharmacologic therapy on changes in aortic aneurysm diameter or on events related to aortic dilation as outcome results.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.1.3 Data Extraction\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data extraction process was carried out by two independent researchers who sought out pertinent articles, screened potential ones based on the set eligibility criteria, and performed data extraction using a standardized datasheet independently. Any disagreement was addressed through discussions involving a third researcher. The extracted data covered the following details: authorship, publication year, country of origin, study type, inclusion criteria, participant numbers, interventions (specifically, metformin), and study results. Treatment strategies for aortic changes in patients with aortic aneurysms, with or without the use of metformin, were compared without any restrictions on treatment history. The main data points extracted included: the first author, year of publication, study design, initial AAA diameter, sample size, follow-up duration, and AAA growth rate. The methodological quality of cohort and case-control studies was evaluated using the Newcastle-Ottawa scale (NOS)\u003csup\u003e[\u003c/sup\u003e\u003csup\u003e2\u003c/sup\u003e\u003csup\u003e0]\u003c/sup\u003e. The two researchers independently completed data extraction and quality assessment, and any discrepancies were resolved through discussion.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.1.4 Quality Assessment\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo assess the quality of the studies, we used the Newcastle-Ottawa Scale (NOS). The NOS checklist has three quality parameters: (1) selection of the study groups; (2) comparability of the groups; and (3) ascertainment of either the exposure or outcome of interest for case-control or cohort studies respectively. Each study received a score ranging from zero to nine. Studies that achieved a score of seven or more were deemed to be of high quality.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.1.5 Statistical Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe statistical analysis was conducted using Review Manager (version 5.4). A meta-analysis was performed to compare AAA growth between the intervention and control groups. The AAA growth (mm/year) in both groups was presented as a mean and standard deviation (SD), and the results were expressed as a mean difference (MD) with its corresponding 95% confidence interval (95% CI)\u003csup\u003e[\u003c/sup\u003e\u003csup\u003e2\u003c/sup\u003e\u003csup\u003e1]\u003c/sup\u003e. Heterogeneity across the studies was quantitatively assessed by the I\u003csup\u003e2\u003c/sup\u003e statistic, with a threshold of 50% indicating significant heterogeneity. If statistical heterogeneity was not present (I\u003csup\u003e2\u003c/sup\u003e \u0026lt; 50%), a fixed effect model was used for the analysis. If the heterogeneity was significant (I\u003csup\u003e2\u003c/sup\u003e \u0026gt; 50%), the random effects model was used, and a sensitivity analysis was performed to pinpoint the source of heterogeneity\u003csup\u003e[\u003c/sup\u003e\u003csup\u003e2\u003c/sup\u003e\u003csup\u003e2]\u003c/sup\u003e. If the cause of heterogeneity could not be determined, we continued with the random-effects model\u003csup\u003e[\u003c/sup\u003e\u003csup\u003e2\u003c/sup\u003e\u003csup\u003e3]\u003c/sup\u003e. Statistical significance was denoted by a \u003cem\u003eP\u003c/em\u003e-value less than 0.05. The results of the analysis were visualized using forest plots.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.1.6 Assessment of Publication Bias\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo evaluate the presence of publication bias in our meta-analysis, we employed funnel plots, a widely recognized method for detecting such biases. These plots are particularly effective in illustrating asymmetry, which often indicates potential publication bias. The funnel plot approach involves plotting the treatment effects estimated from individual studies against a measure of study size or precision, usually the standard error. We created funnel plots for each outcome measure analyzed in our meta-analysis. Visual inspection of these plots was the first step in assessing publication bias. A symmetrical distribution of studies within the funnel plot suggested the absence of publication bias, while asymmetry indicated its potential presence. Asymmetry in funnel plots can also result from other factors such as heterogeneity among study methodologies or true differences in effect sizes. Therefore, the results of the funnel plot analysis should be interpreted with caution, considering other contextual factors related to the studies included in the meta-analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.2 Mendelian randomization analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.2.1 Study Design\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe design of this study unfolds as follows, and is graphically represented in \u003cstrong\u003eFigure 1\u003c/strong\u003e. We utilized summary-level data extracted from Genome-Wide Association Studies (GWAS) to form the foundation of our analysis. Our primary focus in this study is on five genes that are known to be targeted by metformin, a medication widely used in the treatment of type 2 diabetes. These genes, namely MCI, AMPK, GDF15, MG3, and FBP1, were identified and selected based on extensive reviews of existing literature\u003csup\u003e[24, 25]\u003c/sup\u003e. Our aim was to investigate whether there is a direct causal relationship between metformin treatment, the effects of these specific genes, and the incidence of abdominal aortic aneurysm (AAA), and to explore the mediating role of serum lipids, including high-density lipoprotein (HDL), low-density lipoprotein (LDL), triglycerides (TG) and total cholesterol (TC), between metformin treatment and AAA. The mediation effect analysis was completed using the traditional two-step method, and the indirect effect and proportion were obtained using the delta method\u003csup\u003e[26]\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMR is a method that uses genetic variants, or single nucleotide polymorphisms (SNPs), as instrumental variables (IVs) to estimate the causal effect of an exposure (in this case, metformin treatment and its target genes) on an outcome (AAA incidence). This approach is particularly valuable as it helps to overcome the limitations of conventional observational studies, such as confounding and reverse causation. MR analysis is built on three core assumptions. Firstly, we assume that the SNPs we select as IVs are strongly associated with the exposure. This means that these genetic variants should directly influence the action of metformin or its target genes. Secondly, these SNPs must be independent of any confounding factors that might distort the true relationship between exposure and outcome. This is to ensure that any observed associations are not due to these confounding variables, but rather are indicative of a direct causal link. Finally, the third assumption is that the IVs are related to the outcome only via the exposure. In other words, these SNPs should influence AAA incidence only through their effect on metformin treatment and its target genes, and not via any other biological pathways. This assumption is critical to maintain the validity of our MR analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.2.2 Data sources\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data used in this study is drawn entirely from the Integrated Exposure Unit (IEU) Open GWAS project, FinRegistry project and Ensembl genome database project. IEU Open GWAS is a comprehensive summary database that contains information from a multitude of Genome-Wide Association Studies (GWAS). FinRegistry is a joint research project of the Finnish Institute of Health and Welfare (THL) and the Data Science and Genetic Epidemiology Lab research group at the Institute for Molecular Medicine Finland (FIMM), University of Helsinki. The specific characteristics of the datasets used in our study are presented in \u003cstrong\u003eTable 1\u003c/strong\u003e. Our research operates under the ethical approvals granted to the original GWAS studies from which we sourced our data. This means we were not required to obtain new informed consent from patients or to meet any additional ethical requirements. All the datasets we selected to use were derived entirely from European population samples. The first set of data we used pertains to the exposure of individuals to metformin treatment. This data was obtained from the UK Biobank dataset, a resource that includes genetic and health information from a staggering total of 462,933 individuals. Within this dataset, there were 11,552 cases (individuals who had received metformin treatment) and 451,381 controls (those who had not received metformin treatment). Crucially, this classification was based solely on whether or not the individuals had been exposed to metformin, without any consideration of variables such as age or gender. To investigate the exposure of individuals to metformin's target genes, we turned to a different dataset from Ensembl, which includes data from 30,721 individual subjects. Ensembl is a comprehensive database providing detailed information about the genes, variants, and functional interpretations in a wide range of species. The dataset of lipid indexes as mediating factors came from a metabolomics study of 115,082 European individuals published in 2022 by T. G. Richardson et al. The outcome dataset, which includes information on the incidence of Abdominal Aortic Aneurysm (AAA), was sourced from the FinnGen dataset. This dataset includes a total of 473,681 individuals, with a breakdown of 4,217 cases (those diagnosed with AAA) and 469,464 controls (those without AAA). The standard definition code for AAA cases is I71.3 and I71.4 from the International Classification of Diseases 10th revision (ICD-10).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.2.3 Selection of instrumental variables\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe process of selecting instrumental variables (IVs) for our study was involving multiple stages of filtering and testing to ensure the validity of our MR analysis. Our first step was to identify potential IVs among independent genetic variants that exhibited genome-wide significance, defined as having a P-value less than 5e-08. This indicates that there is a less than one in twenty million chance that the observed associations between these genetic variants and the exposure variables occurred by random chance. Thus, selecting SNPs with genome-wide significance adds a layer of robustness to our analysis. After potential IVs were identified, we proceeded to test for linkage disequilibrium (LD) among the selected single nucleotide polymorphisms (SNPs). LD is a phenomenon that occurs when SNPs are inherited together more often than would be expected by chance. This can introduce bias into our analysis, as it could artificially inflate the perceived association between our IVs and the exposure variables. To avoid this, we set stringent LD thresholds: for SNPs related to metformin treatment and phenotypes used as mediators in the two-step analysis, we only retained those with an r2 value less than 0.01 and a physical distance greater than 10000 kilobases (kb); for SNPs related to the target genes, we only retained those with an r2 value greater than 0.1 and a distance greater than 100kb. Next, we discarded any weak IVs, defined as those with an F-statistic less than 10. The F-statistic is a measure of the strength of the instrument, and a low F-statistic can suggest that the IV is weakly correlated with the exposure. In this study, the F statistic was calculated as follows: F = R2(n - k\u0026nbsp;-\u0026nbsp;1)/k (1 - R2), where R2, n, and k denote the proportion of variance in exposure explained by selected genetic tools, the sample size of the exposure GWAS, and the number of chosen genetic tools, respectively\u003csup\u003e[\u003c/sup\u003e\u003csup\u003e27\u003c/sup\u003e\u003csup\u003e]\u003c/sup\u003e. Finally, we harmonized the palindromic SNPs to ensure that the alleles of the IVs in the exposure and outcome datasets had consistent effects. Palindromic SNPs are those where the major and minor alleles cannot be unambiguously determined due to their palindromic nature (A/T or C/G). Detailed information on all used SNPs can be found in \u003cstrong\u003eSupplementary Tables 1-11\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.2.4 Statistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe employed five commonly used Mendelian Randomization (MR) methods, each providing a unique approach to analyze the data. A significance threshold of\u0026nbsp;\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.05 was established for all tests.\u003c/p\u003e\n\u003cp\u003eA. Inverse-Variance Weighted (IVW) Method: This method was the primary MR approach we used to infer causal relationships. The IVW method operates under the assumption that all instrumental variables (IVs) exert a common causal effect on the outcome through the exposure. It aggregates the effect estimates from each IV, using a meta-analysis-like framework. In this framework, the inverse variance of each effect estimate is used as a weight, allowing us to derive a summarized causal estimate. This approach is particularly powerful when the IVs are strong and there is no evidence of horizontal pleiotropy\u003csup\u003e[28, 29]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eB. MR-Egger Method: We used the MR-Egger method specifically to test for horizontal pleiotropy, which occurs when IVs influence the outcome through pathways that are not related to the exposure. This is crucial because such pleiotropy can bias MR estimates. The MR-Egger intercept provides a test for directional pleiotropy; a\u0026nbsp;\u003cem\u003eP\u003c/em\u003e-value of less than 0.05 here suggests significant horizontal pleiotropy\u003csup\u003e[30]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eC. Weighted Median, Weighted Mode, and Simple Mode Methods: These methods offer alternative ways to calculate the causal estimate. They are particularly useful when there is potential invalidity among some IVs, as they can provide more robust estimates under certain conditions of IV invalidity\u003csup\u003e[31\u003c/sup\u003e\u003csup\u003e, 3\u003c/sup\u003e\u003csup\u003e2]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eD. Cochran’s Q Test for Heterogeneity: To assess the heterogeneity among the IV estimates, we employed Cochran’s Q test. A\u0026nbsp;\u003cem\u003eP\u003c/em\u003e-value less than 0.05 in this test indicates the presence of heterogeneity, suggesting variability in the causal effects estimated by different IVs\u003csup\u003e[\u003c/sup\u003e\u003csup\u003e2\u003c/sup\u003e\u003csup\u003e9]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eE. Leave-One-Out Analysis: To ensure the robustness of our findings, we conducted a leave-one-out analysis. This involves sequentially excluding each SNP from the set of IVs and recalculating the causal estimate. This process helps identify any single SNP that might disproportionately influence the results, ensuring the reliability of our findings.\u003c/p\u003e\n\u003cp\u003eAll statistical analyses were conducted using R software (version 4.3.1), with specific analyses carried out using the \"TwoSampleMR\" and \"forestploter\" packages. These tools are specifically designed for MR analyses and provide a comprehensive suite of functions to perform various MR-related statistical tests. This thorough approach to statistical analysis ensures that our results are both reliable and valid, providing a strong foundation for our conclusions regarding the causal relationship between metformin treatment and AAA suppression.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.3 Colocalization analysis study\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eColocalization refers to the spatial overlap of two or more entities within a biological sample, and it can be measured quantitatively using various statistical methods. Colocalization analysis is a statistical method used to test whether two input phenotypes are driven by the same genetic variant site in a certain region, thereby strengthening the evidence of association between the two phenotypes. These phenotypes can be molecular phenotypes, continuous traits or binary traits. Colocalization analysis can identify genes or gene variants that may play a role in multiple traits or diseases, thereby providing new insights into biological mechanisms or providing new targets for disease prevention and treatment. Colocalization analysis relies on four assumptions. For a certain genome interval, the assumptions are as follows:\u003c/p\u003e\n\u003cp\u003eH0: Phenotype 1 and phenotype 2 are not significantly related to all SNP sites in a certain genomic region;\u003c/p\u003e\n\u003cp\u003eH1/H2: Phenotype 1/phenotype 2 is significantly related to the SNP site in a certain genomic region, but phenotype 2/phenotype 1 has nothing to do with it;\u003c/p\u003e\n\u003cp\u003eH3: Phenotype 1 and Phenotype 2 are significantly associated with SNP sites in a certain genomic region, but are driven by two independent causal variant sites;\u003c/p\u003e\n\u003cp\u003eH4: Phenotype 1 and Phenotype 2 are significantly related to SNP sites in a certain genomic region and are driven by the same causal variant site.\u003c/p\u003e"},{"header":"3. Results","content":"\u003cp\u003e\u003cstrong\u003e3.1 Study Inclusion for Meta-analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur study involved a comprehensive review and meta-analysis of the existing literature related to our research question. This process was characterized by an exhaustive search, rigorous screening, and the eventual inclusion of high-quality studies, as illustrated in \u003cstrong\u003eFigure\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e2\u003c/strong\u003e. The literature search yielded 54 relevant articles in total. To ensure the validity and applicability of our study, we set specific inclusion and exclusion criteria for the articles. These criteria were meticulously designed based on the scope of our research, the quality of the studies, and the relevance of their findings to our research question. Each of the 54 articles underwent a thorough screening process. This involved examining the articles' titles and abstracts, reading the full texts where necessary, and evaluating their methodology and findings against our inclusion and exclusion criteria. This rigorous scrutiny ensured that only the most pertinent and high-quality studies were included in our review. After this rigorous screening process, a total of five articles(10-14) were deemed suitable for inclusion in our review. These five articles, in turn, comprised a total of 8 cohort studies, involving an impressive total of 14710 participants. The scale of these studies enhances the robustness and generalizability of our review's findings. The characteristics of each included study, such as the study design, sample size, population characteristics, exposure and outcome measures, and key findings, are presented in \u003cstrong\u003eTable 2\u003c/strong\u003e. This table provides a snapshot of the breadth and depth of the studies included in our review, offering insights into the diverse methodologies and findings from which our conclusions are drawn. Importantly, the quality of the included studies was assessed using the Newcastle-Ottawa Scale (NOS), a widely recognized tool for evaluating the quality of non-randomized studies in meta-analyses. All seven articles scored greater than 7 on the NOS scale, indicating high quality.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.2 Meta-analysis for association between metformin and enlargement of abdominal aortic diameter\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFirstly, this substantial sample size was split into two groups: one group of 598 participants had been prescribed metformin, while the other group of 8730 participants had not been prescribed metformin. Secondly, this information was used to categories patients into two groups: patients with no history or medically recorded diagnosis of diabetes (n = 1820) and patients previously diagnosed with diabetes prescribed metformin at the time of recruitment (n = 238). In \u003cstrong\u003eFigure\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003cstrong\u003eA\u003c/strong\u003e, our meta-analysis performed between diabetes prescribed metformin compared to diabetes not prescribed metformin and the progression of abdominal aortic diameter enlargement. Our forest plots, a graphical display designed to illustrate the relative strength of treatment effects in multiple studies, showed that metformin usage was associated with an inhibition of abdominal aortic diameter enlargement when compared to patients not treated with metformin. This was shown through a Mean Difference (MD) of -0.60 (95%CI: -0.97\u0026nbsp;-\u0026nbsp;-0.23, \u003cem\u003eP\u003c/em\u003e = 0.002), indicating a significantly lesser enlargement in the group treated with metformin. Although there is no publication bias (\u003cstrong\u003eFigure\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003cstrong\u003eB\u003c/strong\u003e), this high level of heterogeneity (I\u003csup\u003e2\u003c/sup\u003e statistic of 80% and P\u0026lt;0.0001) suggests that the studies varied considerably in their outcomes, which may reflect differences in study design, population characteristics, or other factors. In response to the high heterogeneity, we conducted a sensitivity analysis to identify potential sources of these issues. Following the exclusion of one cohort of the study by Golledge et al. (2019), patients with initial AAA diameter≤50 mm, the heterogeneity did not decrease (I\u003csup\u003e2\u003c/sup\u003e=83%, P\u0026lt;0.00001), as shown in \u003cstrong\u003eFigure\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e2\u003c/strong\u003e. This was shown through MD=-0.61 (95%CI: -1.05\u0026nbsp;-\u0026nbsp;-0.16, \u003cem\u003eP\u003c/em\u003e =\u0026nbsp;0.008, \u003cstrong\u003eFigure\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003cstrong\u003eC\u003c/strong\u003e), indicating a significantly lesser enlargement in the group treated with metformin, also there is no publication bias (\u003cstrong\u003eFigure\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003cstrong\u003eD\u003c/strong\u003e). Finally, our study compared diabetes prescribed metformin patients with no history or medically recorded diagnosis of diabetes patients, showed that metformin usage was associated with an inhibition of abdominal aortic diameter enlargement (MD=-1.00, 95%CI: -1.33\u0026nbsp;-\u0026nbsp;-0.68, \u003cem\u003eP\u003c/em\u003e \u0026lt;\u0026nbsp;0.000001, \u003cstrong\u003eFigure\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003cstrong\u003eE\u003c/strong\u003e). Inversely, this meta-analysis, there is no significant heterogeneity was observed (I\u003csup\u003e2\u003c/sup\u003e =\u0026nbsp;24%, \u003cem\u003eP\u003c/em\u003e=\u0026nbsp;0.27) as well as publication bias (\u003cstrong\u003eFigure\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003cstrong\u003eF\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.3 Causal Effects of Metformin Treatment on AAA\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe results of the Mendelian randomization (MR) analysis, summarized in \u003cstrong\u003eSupplementary Table 8\u003c/strong\u003e, showed that the slope of the solid line in the scatter plot of genetic variation on metformin treatment and AAA incidence (\u003cstrong\u003eFigure 4\u003c/strong\u003e\u003cstrong\u003eA\u003c/strong\u003e) indicated a negative correlation between metformin treatment and AAA. Single SNP analysis also supported this result overall (\u003cstrong\u003eFigure 4\u003c/strong\u003e\u003cstrong\u003eD\u003c/strong\u003e). All five analyses yielded negative beta values. The IVW method demonstrated a negative association between metformin treatment and AAA (OR = 0.0108, 95%CI = 0.000204 to 0.572,\u0026nbsp;\u003cem\u003eP\u003c/em\u003e = 0.0254). The weighted mode analysis also produced consistent results\u0026nbsp;(OR = 0.0024,\u0026nbsp;95% CI =\u0026nbsp;0.000008 to\u0026nbsp;0.735,\u0026nbsp;\u003cem\u003eP\u003c/em\u003e = 0.0451). However, it is important to note that both the weighted mean and simple mode methods did not yield significant positive results.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.4 Causal Effects of Metformin Drug Target Genes on AAA\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo further investigate the potential causal effects of metformin-related genes on AAA, we conducted MR analyses on 40 genes associated with the 5 target genes of metformin. Using a significance threshold of\u0026nbsp;\u003cem\u003eP\u003c/em\u003e-value \u0026lt; 0.05, and taking the IVW method as the primary method, we identified evidence suggesting a decreased risk of AAA associated with increased expression of MCI-related genes NDUFAF1, NDUFAF3, NDUFB5, NDUFB9, NDUFB10, NDUFS5, and NDUFV2, AMPK-related gene PRKAG3, and MG3-related gene GPD2. Among them, the expression of NDUFAF1, NDUFAF3, NDUFB9, NDUFB10, NDUFV2, PRKAG3 and GPD2 is negatively correlated with the prevalence of AAA. On the other hand, the expression of NDUFB5 and NDUFS5 is positively correlated with the prevalence of AAA. A summary of the analysis results for all positive genes can be found in \u003cstrong\u003eSupplementary Table 9\u003c/strong\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.5 Sensitivity Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe conducted multiple sensitivity analyses to assess the stability and reliability of our findings. Cochran's Q test, used to evaluate heterogeneity, revealed some heterogeneity between metformin treatment and AAA SNPs (Q = 73.623,\u0026nbsp;\u003cem\u003eP\u003c/em\u003e = 0.018, \u003cstrong\u003eSupplementary\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;Table\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;8B\u003c/strong\u003e), while no heterogeneity was found between target gene SNPs and AAA. The symmetry of the funnel plot also indicated consistent results (\u003cstrong\u003eFigure 4\u003c/strong\u003e\u003cstrong\u003eB\u003c/strong\u003e). MR-Egger regression test, used to test for horizontal pleiotropy, showed no evidence of horizontal pleiotropy (MR-Egger intercept = 0.0092; SE = 0.016;\u0026nbsp;\u003cem\u003eP\u003c/em\u003e = 0.564). Additionally, it is worth noting that leave-one-out sensitivity testing suggested a strong influence of SNP rs34872471 on the causal effect of metformin treatment on AAA (\u003cstrong\u003eFigure 4\u003c/strong\u003e\u003cstrong\u003eC\u003c/strong\u003e). Overall, our conclusions regarding the causal relationship between metformin treatment and AAA are stable and reliable (\u003cstrong\u003eFigure 5\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.6 Mediating effect analysis of serum lipids index\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBased on a two-step mediation analysis, we evaluated the potential mediating effects of HDL, LDL, TC and TG in the previously determined two-sample MR stages. The initial test confirmed the causal relationship between metformin treatment and TC, LDL, and between TC, LDL and AAA, but there was no significant causal association between metformin treatment and HDL, TG. This made it possible to study the mediating effects of TC and LDL. Subsequently, we assessed the mediation effects of TC and LDL. All mediation Mendelian randomization analysis used IVW as the primary method, the results are revealed in \u003cstrong\u003eTable 3\u003c/strong\u003e. For LDL, the mediation effect ratio (IE_div_TE) was 28.84% with an OR of 0.271 (95% CI = 0.133-0.552); for TC, IE_div_TE was 25.17% with an OR of 0.3199 (95% CI = 0.159-0.644), indicating that there was a mediation effect between metformin and AAA mediated by LDL or TC. \u003cstrong\u003eSupplementary Table 10\u003c/strong\u003e provides full results of the two-step analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.7 Colocalization analysis results of all nine positive genes\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTaking the\u0026nbsp;\u003cem\u003eP\u003c/em\u003e-value\u0026gt;0.75 of PP.H4 as the positive SNP site as the benchmark, the graphs drawn from the analysis results of all positive genes are shown in\u0026nbsp;\u003cstrong\u003eFigure 6\u003c/strong\u003e. For the nine genes with positive results in the Mendelian randomization analysis (GPD2, NDUFAF1, NDUFAF3, NDUFB5, NDUFB9, NDUFB10, NDUFS5, NDUFV2 and PRKAG3), the SNPs with the strongest causal associations are rs2711744, rs62019915, rs13010713, rs79975467, rs10745332, rs11761528, rs77992778, rs1442685 and rs1365963. Among these nine genes, GPD2 has significantly more positive SNPs, which indicates that GPD2 may be the most important site of action of metformin in inhibiting the onset of AAA. In addition, it is worth noting that the analysis results of NDUFS5 and NDUFV2 showed almost no SNPs that met the screening criteria.\u003c/p\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eAbdominal aortic aneurysm (AAA) remains a significant silent player in the pathogenesis of cardiovascular diseases, often going undiagnosed until rupture has occurred, leading to high mortality rates. Currently, the therapeutic interventions for AAA are majorly restricted to surgical approaches like open repair and endovascular aneurysm repair\u003csup\u003e[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]\u003c/sup\u003e. These techniques mainly focus on the symptomatic stages of AAA when the aneurysm has enlarged beyond a particular threshold. What is more concerning is that these surgical procedures accompany their fair share of complications such as graft infections, endoleaks, and renal impairment making the management of AAA challenging to clinicians\u003csup\u003e[\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]\u003c/sup\u003e. Despite the rigorous research in the field of therapeutic methods, there is much left undiscovered and unexplored, especially when considering pharmaceutical treatments for AAA. Additionally, several trials have been performed to repurpose drugs like doxycycline and roxithromycin, suggesting potential benefits in AAA. Still, these results are yet to translate clinically due to insufficient evidence\u003csup\u003e[\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]\u003c/sup\u003e. Furthermore, the use of other drugs, like losartan, has revealed disappointing results with no significant impact on AAA progression. Our exploration into the realm of metformin\u0026rsquo;s potential effect on AAA has led us through a broad range of methodologies, amplified our understanding of the connection between these two diverse yet integrally connected areas, and opened gateways to new avenues of research. The potential transformative implication of our study goes beyond the traditional clinical context and has significant echos in the research and policy arenas. The experimental evidence obtained through rodent models, though useful in deciphering mechanistic insights, often falls short in their translatability to humans due to differences in physiology, anatomy and disease mechanisms\u003csup\u003e[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]\u003c/sup\u003e. This gap provides an impetus for researchers to devise human-centered, clinically relevant research models that better mimic human disease paths. The challenges in uncovering genetic complexities, disease pathways, and treatment effects emphasize the need for the development of novel tools and approaches in predictive modeling.\u003c/p\u003e \u003cp\u003eOur meta-analysis exhibits varying findings on the association between metformin and AAA. Some studies imply a protective effect of metformin against AAA, others provide inconclusive or even contrasting results\u003csup\u003e[\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]\u003c/sup\u003e. These discrepancies underline the multifactorial nature of AAA, influenced not only by metformin but also by a host of confounding factors such as medical history, lifestyle, and genetic predisposition. Therefore, although our results suggest a potential correlation, they call for a cautious interpretation and advocate for a need for more rigorous to establish a precise causal relationship conclusively. It's worth noting that the Mendelian Randomization comes with its own set of assumptions, which if violated, can lead to biased results. Therefore, it\u0026rsquo;s crucial to conduct sensitivity analysis and use various MR methods to improve the robustness of our findings. Our study represents a cautious but forward step in this direction. There is an essential need for more research and replication of our findings in a wider range of populations with varying genetic backgrounds and lifestyle factors, to improve the generalizability of our conclusions\u003csup\u003e[\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eMechanically, our Mendelian randomization analysis has provided significant mechanistic insights into the potential action of metformin on AAA. AAA embodies a complex etiology involving matrix degradation, inflammation, and oxidative stress, making the development of an optimal pharmaceutical intervention challenging\u003csup\u003e[\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]\u003c/sup\u003e. A promising avenue for cultivating novel AAA treatments involves gaining a deeper understanding of the specific genetic and molecular insurgents involved in AAA development and progression. Such insights can guide the research of new drugs targeting these specific mechanisms, paving the way for personalized medicine in AAA management. It appears that metformin potentially impacts AAA through several target genes such as AMPK, GDF15, MG53, FBP1, and MCI. AMPK (AMP-activated protein kinase), activated by metformin, is known for its role in energy metabolism and appears to be protective against AAA by triggering vasodilation, reducing vascular inflammation, and inhibiting vascular smooth muscle cell proliferation\u003csup\u003e[\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]\u003c/sup\u003e. GDF15 (Growth Differentiation Factor 15) and MG53 (Mitsugumin 53), both upregulated by metformin, contribute to the control of metabolic dysfunction, inflammation, and tissue repair\u003csup\u003e[\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]\u003c/sup\u003e. These processes are key components in the development and progression of AAA, thus highlighting their potential role in AAA prevention. FBP1 (Fructose-1,6-Bisphosphatase 1), a gene also targeted by metformin, is one of the essential regulatory enzymes in gluconeogenesis. Its regulation via metformin could have cardio-protective effects by increasing glucose uptake and reducing glucose production. Repurposing FBP1: dephosphorylating IκBα to suppress NF-κB\u003csup\u003e[\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e]\u003c/sup\u003e. Activation of NF-κB within endothelial cells leads to an increase in the expression of adhesion molecules. This can initiate the infiltration of macrophages and subsequent inflammation within the adventitia and media layers in AAA. Therefore, the endothelium is significantly involved in vascular remodeling and the development of aneurysms via its internal NF-κB signaling pathways\u003csup\u003e[\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]\u003c/sup\u003e. Finally, MCI (Metabolic Complication Indicator), targeted by metformin, plays a role in signaling metabolic abnormalities, highlighting a possible role for metformin in mitigating AAA through metabolic regulation. In addition, according to further two-step mediation analysis, the key way for metformin to reduce the risk of AAA by regulating blood lipid levels should be to protect vascular smooth muscle cells from cholesterol-induced functional changes by lowering cholesterol levels, especially low-density lipoprotein cholesterol levels\u003csup\u003e[\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]\u003c/sup\u003e. The pathway by which metformin regulates cholesterol levels is AMPK/SIRT1\u003csup\u003e[\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]\u003c/sup\u003e. However, despite these compelling mechanistic insights, more definitive evidence is needed to substantiate these associations. Hence, further research should focus on validating these target effects in experimental models and clinical trials, thereby providing a more in-depth understanding of the mechanisms underlying these associations. This approach would contribute significantly to the development of potential therapeutics aiming to leverage metformin\u0026rsquo;s impact on AAA.\u003c/p\u003e \u003cp\u003eStrengths and limitations that come to light through the course of our research. Starting with strengths, our Mendelian randomization approach allows us to bypass some of the inherent challenges in conducting RCTs. This gives our study a significant advantage as it offers a relatively unbiased estimation of the causal effect - a critical stepping stone in the therapeutic investigation. Using genetic variants as instrumental variables allows us to estimate the effects of metformin therapy on AAA, providing us substantial evidence which otherwise would have been challenging to obtain. By blending an extensive literature search and synthesis, this approach ensures comprehensive coverage of existing evidence, minimizing the possibility of missing relevant information, and providing a thorough understanding of the research landscape surrounding metformin and AAA. The regulation of serum lipids can also be considered as a potential mechanism of action for the genetic causal association between metformin and AAA. However, there are also limitations to be acknowledged. While MR studies mimic the randomizing nature of RCTs, they are not entirely free of biases. It's noteworthy that MR relies heavily on the assumptions it makes about the genetic variants used. Any violation of these assumptions can introduce biases that may over or underestimate the true effect. Furthermore, despite the comprehensive literature search in the meta-analysis phase, the possibility of oversight of unpublished studies and articles not indexed in the database may introduce publication bias. Lastly, while we have pinpointed potential targets, the biological pathways involving metformin and AAA are highly complex and our understanding is still at a nascent stage. Our study proposes a pathway of investigation and does not establish a definitive cause-and-effect relationship.\u003c/p\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eThis study provides compelling evidence that metformin, traditionally used for type II diabetes, holds promise as a therapeutic agent for Abdominal Aortic Aneurysm (AAA). Our findings, derived from meta-analysis, Mendelian randomization, and colocalization analysis, suggest that metformin can significantly inhibit AAA progression. Notably, colocalization analysis identified specific SNPs in genes such as GPD2 that are associated with both metformin targets and AAA, underscoring a genetic basis for this therapeutic effect. Furthermore, the mediating effect of metformin on lipid profiles, particularly its ability to lower LDL and total cholesterol levels, highlights an important pathway through which metformin may exert its protective effects. These insights advocate for further clinical trials to validate metformin's efficacy in AAA treatment, potentially offering a non-surgical option for managing this life-threatening condition.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Fundamental Research Funds for the Central Universities (grant number: DUT22YG107), the National Natural Science Foundation of China (grant number: 81600370), the China Postdoctoral Science Foundation (grant number: 2018M640270) and the Natural Science Foundation of Liaoning Province (2023-MS-096) for Yanshuo Han, also supported by National Natural Science Foundation of China (grant number: 81970402 and 82170507) for Jian Zhang, and supported by International Science and Technology Cooperation Program of Liaoning Province (grant number: 2023JH2/10700019).\u003c/p\u003e\n\u003cp\u003eWe would like to acknowledge all the above institutions that provided financial support. In addition, We would like to acknowledge the patients and medical institutions who selflessly contributed valuable publicly available data to this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e6. Declaration of financial/other relationships\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e7. Author contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eZ.Z. found the data, performed the MR analysis and interpreted all sections regarding the methods and results of the MR analysis. Y.H., J.U. and A.H. designed the study and drafted the manuscript. J.W., X.Z., D.J., Y.L., H.J. and J.Z. contributed other parts including bibliometric and meta-analysis, and checked and revised the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e8. Ethical approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEach cohort included in this study has been conducted using published studies and consortia providing publicly \u0026nbsp; available \u0026nbsp;summary \u0026nbsp;statistics, so no additional ethical approval was required.. \u0026nbsp;All \u0026nbsp;original \u0026nbsp; studies \u0026nbsp;has received \u0026nbsp;ethical \u0026nbsp; approval \u0026nbsp;and \u0026nbsp;agreed \u0026nbsp; to participate, \u0026nbsp;and \u0026nbsp;summary-level \u0026nbsp; data \u0026nbsp;were provided for analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e9. Data availability statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll of the data generated or analyzed in this study are either contained in the article/supplementary material or in the data repositories listed in References. Further inquiries can be directed to the corresponding authors.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAhmed Z, Zeeshan S, Mendhe D, Dong X. Human gene and disease associations for clinical-genomics and precision medicine research. Clin Transl Med. 2020;10(1):297-318.\u003c/li\u003e\n \u003cli\u003eBaman JR, Eskandari MK. What Is an Abdominal Aortic Aneurysm? JAMA. 2022;328(22):2280.\u003c/li\u003e\n \u003cli\u003eHan Y, Zhang S, Zhang J, Ji C, Eckstein HH. Outcomes of Endovascular Abdominal Aortic Aneurysm Repair in Octogenarians: Meta-analysis and Systematic Review. Eur J Vasc Endovasc Surg. 2017;54(4):454-63.\u003c/li\u003e\n \u003cli\u003eSiordia JA. Beta-Blockers and Abdominal Aortic Aneurysm Growth: A Systematic Review and Meta-Analysis. Curr Cardiol Rev. 2021;17(4):e230421187502.\u003c/li\u003e\n \u003cli\u003eHosseini A, Sahranavard T, Reiner Z, Jamialahmadi T, Dhaheri YA, Eid AH, et al. Effect of statins on abdominal aortic aneurysm. Eur J Pharm Sci. 2022;178:106284.\u003c/li\u003e\n \u003cli\u003eThatcher SE, Zhang X, Howatt DA, Yiannikouris F, Gurley SB, Ennis T, et al. Angiotensin-converting enzyme 2 decreases formation and severity of angiotensin II-induced abdominal aortic aneurysms. Arterioscler Thromb Vasc Biol. 2014;34(12):2617-23.\u003c/li\u003e\n \u003cli\u003eGoldberg RB, Orchard TJ, Crandall JP, Boyko EJ, Budoff M, Dabelea D, et al. Effects of Long-term Metformin and Lifestyle Interventions on Cardiovascular Events in the Diabetes Prevention Program and Its Outcome Study. Circulation. 2022;145(22):1632-41.\u003c/li\u003e\n \u003cli\u003eLi JZ, Li YR. Cardiovascular Protection by Metformin: Latest Advances in Basic and Clinical Research. Cardiology. 2023;148(4):374-84.\u003c/li\u003e\n \u003cli\u003eTai S, Sun J, Zhou Y, Zhu Z, He Y, Chen M, et al. Metformin suppresses vascular smooth muscle cell senescence by promoting autophagic flux. J Adv Res. 2022;41:205-18.\u003c/li\u003e\n \u003cli\u003eFujimura N, Xiong J, Kettler EB, Xuan H, Glover KJ, Mell MW, et al. Metformin treatment status and abdominal aortic aneurysm disease progression. J Vasc Surg. 2016;64(1):46-54 e8.\u003c/li\u003e\n \u003cli\u003eGolledge J, Morris DR, Pinchbeck J, Rowbotham S, Jenkins J, Bourke M, et al. Editor\u0026apos;s Choice - Metformin Prescription is Associated with a Reduction in the Combined Incidence of Surgical Repair and Rupture Related Mortality in Patients with Abdominal Aortic Aneurysm. Eur J Vasc Endovasc Surg. 2019;57(1):94-101.\u003c/li\u003e\n \u003cli\u003eGolledge J, Moxon J, Pinchbeck J, Anderson G, Rowbotham S, Jenkins J, et al. Association between metformin prescription and growth rates of abdominal aortic aneurysms. Br J Surg. 2017;104(11):1486-93.\u003c/li\u003e\n \u003cli\u003eItoga NK, Rothenberg KA, Suarez P, Ho TV, Mell MW, Xu B, et al. Metformin prescription status and abdominal aortic aneurysm disease progression in the U.S. veteran population. J Vasc Surg. 2019;69(3):710-6 e3.\u003c/li\u003e\n \u003cli\u003eUnosson J, Wagsater D, Bjarnegard N, De Basso R, Welander M, Mani K, et al. Metformin Prescription Associated with Reduced Abdominal Aortic Aneurysm Growth Rate and Reduced Chemokine Expression in a Swedish Cohort. Ann Vasc Surg. 2021;70:425-33.\u003c/li\u003e\n \u003cli\u003eHariton E, Locascio JJ. Randomised controlled trials - the gold standard for effectiveness research: Study design: randomised controlled trials. BJOG. 2018;125(13):1716.\u003c/li\u003e\n \u003cli\u003eSchooling CM, Freeman G, Cowling BJ. Mendelian randomization and estimation of treatment efficacy for chronic diseases. Am J Epidemiol. 2013;177(10):1128-33.\u003c/li\u003e\n \u003cli\u003eSanderson E, Glymour MM, Holmes MV, Kang H, Morrison J, Munafo MR, et al. Mendelian randomization. Nat Rev Methods Primers. 2022;2.\u003c/li\u003e\n \u003cli\u003eDavey Smith G, Hemani G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet. 2014;23(R1):R89-98.\u003c/li\u003e\n \u003cli\u003ePage MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71.\u003c/li\u003e\n \u003cli\u003eStang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol. 2010;25(9):603-5.\u003c/li\u003e\n \u003cli\u003eDerSimonian R, Laird N. Meta-analysis in clinical trials revisited. Contemp Clin Trials. 2015;45(Pt A):139-45.\u003c/li\u003e\n \u003cli\u003eCordero CP, Dans AL. Key concepts in clinical epidemiology: detecting and dealing with heterogeneity in meta-analyses. J Clin Epidemiol. 2021;130:149-51.\u003c/li\u003e\n \u003cli\u003eKriston L. Dealing with clinical heterogeneity in meta-analysis. Assumptions, methods, interpretation. Int J Methods Psychiatr Res. 2013;22(1):1-15.\u003c/li\u003e\n \u003cli\u003eZheng J, Xu M, Walker V, Yuan J, Korologou-Linden R, Robinson J, et al. Evaluating the efficacy and mechanism of metformin targets on reducing Alzheimer\u0026rsquo;s disease risk in the general population: a Mendelian randomisation study. Diabetologia. 2022;65(10):1664-75.\u003c/li\u003e\n \u003cli\u003eZheng J, Xu M, Yang Q, Hu C, Walker V, Lu J, et al. Efficacy of metformin targets on cardiometabolic health in the general population and non-diabetic individuals: a Mendelian randomization study. EBioMedicine. 2023;96:104803.\u003c/li\u003e\n \u003cli\u003eCarter AR, Gill D, Davies NM, et al. Understanding the consequences of education inequality on cardiovascular disease: mendelian randomisation study. BMJ. 2019;365:l1855. Published 2019 May 22. doi:10.1136/bmj.l1855\u003c/li\u003e\n \u003cli\u003eBurgess S, Thompson SG. Avoiding bias from weak instruments in Mendelian randomization studies. Int J Epidemiol. 2011;40(3):755\u0026ndash;64.\u003c/li\u003e\n \u003cli\u003eBurgess S, Scott RA, Timpson NJ, Davey Smith G, Thompson SG, Consortium E-I. Using published data in Mendelian randomization: a blueprint for efficient identification of causal risk factors. Eur J Epidemiol. 2015;30(7):543-52.\u003c/li\u003e\n \u003cli\u003eBurgess S, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol. 2013;37(7):658-65.\u003c/li\u003e\n \u003cli\u003eBowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol. 2015;44(2):512-25.\u003c/li\u003e\n \u003cli\u003eBowden J, Davey Smith G, Haycock PC, Burgess S. Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median Estimator. Genet Epidemiol. 2016;40(4):304-14.\u003c/li\u003e\n \u003cli\u003eHartwig FP, Davey Smith G, Bowden J. Robust inference in summary data Mendelian randomization via the zero modal pleiotropy assumption. Int J Epidemiol. 2017;46(6):1985-98.\u003c/li\u003e\n \u003cli\u003eSoo Hoo AJ, Fitzgibbon JJ, Hussain MA, Scully RE, Servais AB, Nguyen LL, et al. Contemporary indications for open abdominal aortic aneurysm repair in the endovascular era. J Vasc Surg. 2022;76(4):923-31 e1.\u003c/li\u003e\n \u003cli\u003eDeery SE, Schermerhorn ML. Open versus endovascular abdominal aortic aneurysm repair in Medicare beneficiaries. Surgery. 2017;162(4):721-31.\u003c/li\u003e\n \u003cli\u003eNiaz OS, Rao A, Abidia A, Parrott R, Refson J, Somaiya P. Surgical and medical interventions for abdominal aortic graft infections. Cochrane Database Syst Rev. 2020;8(8):CD013469.\u003c/li\u003e\n \u003cli\u003eSterpetti AV. Doxycycline and Growth of Abdominal Aortic Aneurysms. JAMA. 2020;324(15):1568-9.\u003c/li\u003e\n \u003cli\u003eWang Z, Guo J, Han X, Xue M, Wang W, Mi L, et al. Metformin represses the pathophysiology of AAA by suppressing the activation of PI3K/AKT/mTOR/autophagy pathway in ApoE(-/-) mice. Cell Biosci. 2019;9:68.\u003c/li\u003e\n \u003cli\u003eGuo J, Wang Z, Xue M, Mi L, Zhao M, Ma C, et al. Metformin protects against abdominal aortic aneurysm by Atg7-induced autophagy. Adv Clin Exp Med. 2022;31(1):59-69.\u003c/li\u003e\n \u003cli\u003eYuan Z, Heng Z, Lu Y, Wei J, Cai Z. The Protective Effect of Metformin on Abdominal Aortic Aneurysm: A Systematic Review and Meta-Analysis. Front Endocrinol (Lausanne). 2021;12:721213.\u003c/li\u003e\n \u003cli\u003eLiu H, Wu W, Xiang W, Yuan J. Lifestyle factors, metabolic factors and socioeconomic status for pelvic organ prolapse: a Mendelian randomization study. Eur J Med Res. 2023;28(1):183.\u003c/li\u003e\n \u003cli\u003eWu Y, Jiang D, Zhang H, Yin F, Guo P, Zhang X, et al. N1-Methyladenosine (m1A) Regulation Associated With the Pathogenesis of Abdominal Aortic Aneurysm Through YTHDF3 Modulating Macrophage Polarization. Front Cardiovasc Med. 2022;9:883155.\u003c/li\u003e\n \u003cli\u003eStepien KL, Bajdak-Rusinek K, Fus-Kujawa A, Kuczmik W, Gawron K. Role of Extracellular Matrix and Inflammation in Abdominal Aortic Aneurysm. Int J Mol Sci. 2022;23(19).\u003c/li\u003e\n \u003cli\u003eHe J, Li N, Fan Y, Zhao X, Liu C, Hu X. Metformin Inhibits Abdominal Aortic Aneurysm Formation through the Activation of the AMPK/mTOR Signaling Pathway. J Vasc Res. 2021;58(3):148-58.\u003c/li\u003e\n \u003cli\u003eWang S, Zhang C, Zhang M, Liang B, Zhu H, Lee J, et al. Activation of AMP-activated protein kinase alpha2 by nicotine instigates formation of abdominal aortic aneurysms in mice in vivo. Nat Med. 2012;18(6):902-10.\u003c/li\u003e\n \u003cli\u003eSanchez-Infantes D, Nus M, Navas-Madronal M, Fite J, Perez B, Barros-Membrilla AJ, et al. Oxidative Stress and Inflammatory Markers in Abdominal Aortic Aneurysm. Antioxidants (Basel). 2021;10(4).\u003c/li\u003e\n \u003cli\u003eCui KM, Hu ZP, Wang YL. MG53 represses high glucose-induced inflammation and angiogenesis in human retinal endothelial cells by repressing the EGR1/STAT3 axis. Immunopharmacol Immunotoxicol. 2022;44(4):484-91.\u003c/li\u003e\n \u003cli\u003eZhang G, Tao J, Lin L, Qiu W, Lu Z. Repurposing FBP1: dephosphorylating IkappaBalpha to suppress NFkappaB. Cell Res. 2023;33(6):419-20.\u003c/li\u003e\n \u003cli\u003eSaito T, Hasegawa Y, Ishigaki Y, Yamada T, Gao J, Imai J, et al. Importance of endothelial NF-kappaB signalling in vascular remodelling and aortic aneurysm formation. Cardiovasc Res. 2013;97(1):106-14.February 2006.\u003c/li\u003e\n \u003cli\u003eSolym\u0026aacute;r M, Ivic I, P\u0026oacute;t\u0026oacute; L, et al. Metformin induces significant reduction of body weight, total cholesterol and LDL levels in the elderly - A meta-analysis. PLoS One. 2018;13(11):e0207947. Published 2018 Nov 26. doi:10.1371/journal.pone.0207947\u003c/li\u003e\n \u003cli\u003eXing H, Liang C, Wang C, Xu X, Hu Y, Qiu B. Metformin mitigates cholesterol accumulation via the AMPK/SIRT1 pathway to protect osteoarthritis chondrocytes. Biochem Biophys Res Commun. 2022;632:113-121. doi:10.1016/j.bbrc.2022.09.074\u003c/li\u003e\n \u003cli\u003eAli A, Unnikannan H, Shafarin J, et al. Metformin enhances LDL-cholesterol uptake by suppressing the expression of the pro-protein convertase subtilisin/kexin type 9 (PCSK9) in liver cells. Endocrine. 2022;76(3):543-557. doi:10.1007/s12020-022-03022-x\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable 1\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eSource of GWAS data for metformin treatment and AAA and eQTL data for five metformin targets\u003c/p\u003e\n\u003cdiv align=\"Left\"\u003e\n \u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"733\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePhenotype\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 46px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eYear\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAuthor\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSamplesize or case/control\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAncestry\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 361px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eWeb Source if Publicly Available\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 733px;\"\u003e\n \u003cp\u003eExposure\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003eMetformin\u003c/p\u003e\n \u003cp\u003etreatment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 46px;\"\u003e\n \u003cp\u003e2018\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003eBen Elsworth\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 103px;\"\u003e\n \u003cp\u003e11,552/451,381\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003eEuropean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 361px;\"\u003e\n \u003cp\u003ehttps://gwas.mrcieu.ac.uk/datasets/ukb-b-14609\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003eTarget Genes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 46px;\"\u003e\n \u003cp\u003e2018\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003eUrmo Vosa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 103px;\"\u003e\n \u003cp\u003e31,684\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003eEuropean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 361px;\"\u003e\n \u003cp\u003ehttps://gwas.mrcieu.ac.uk/datasets/eqtl-a-(Corresponding ENSG ID), Please refer to: https://useast.ensembl.org/index.html\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 733px;\"\u003e\n \u003cp\u003eMediators\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003eLDL-C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 46px;\"\u003e\n \u003cp\u003e2022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003eTom\u0026nbsp;Richardson\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 103px;\"\u003e\n \u003cp\u003e115,082\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003eEuropean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 361px;\"\u003e\n \u003cp\u003ehttps://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST90092883/\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003eHDL-C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 46px;\"\u003e\n \u003cp\u003e2022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003eTom\u0026nbsp;Richardson\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 103px;\"\u003e\n \u003cp\u003e115,082\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003eEuropean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 361px;\"\u003e\n \u003cp\u003ehttps://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST90092822/\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003eTG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 46px;\"\u003e\n \u003cp\u003e2022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003eTom\u0026nbsp;Richardson\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 103px;\"\u003e\n \u003cp\u003e115,082\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003eEuropean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 361px;\"\u003e\n \u003cp\u003ehttps://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST90092992/\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003eTC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 46px;\"\u003e\n \u003cp\u003e2022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003eTom\u0026nbsp;Richardson\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 103px;\"\u003e\n \u003cp\u003e115,082\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003eEuropean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 361px;\"\u003e\n \u003cp\u003ehttps://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST90092985/\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\" style=\"width: 733px;\"\u003e\n \u003cp\u003eOutcome\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003eAbdominal aortic aneurysm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 46px;\"\u003e\n \u003cp\u003e2021\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 103px;\"\u003e\n \u003cp\u003e4,217/469,464\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003eEuropean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 361px;\"\u003e\n \u003cp\u003ehttps://risteys.finregistry.fi/endpoints/I9_ABAORTANEUR\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2\u003c/strong\u003e Summary of Included Studies\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" align=\"left\" width=\"134%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003eAuthors\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003eYear\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003eCountry\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003eStudy design\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 5px;\"\u003e\n \u003cp\u003eCases\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003eParticipants\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003eDrug\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003eFollow-up, Years\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 19px;\"\u003e\n \u003cp\u003eOutcomes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003eNOS\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003eFujimura Naoki\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e2016\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003eUnited States\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003eCohort study\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 5px;\"\u003e\n \u003cp\u003e58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003eElderly patients with diabetes with untreated AAA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e1. Metformin (T2DM);\u003c/p\u003e\n \u003cp\u003e2. Without metformin (T2DM)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003e2.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 19px;\"\u003e\n \u003cp\u003eMaximum aortic diameter growth measured by contrast-enhanced CT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003eGolledge Jonathan\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e2019\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003eAustralia\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003eCohort study A\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 5px;\"\u003e\n \u003cp\u003e1,080\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003ePatients with asymptomatic unrepaired AAA of any diameter\u0026ge;30mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e1. Diabetes with no history or medically recorded；\u003c/p\u003e\n \u003cp\u003e2. Diabetes prescribed metformin at the time of recruitment;\u003c/p\u003e\n \u003cp\u003e3. Diabetes not prescribed metformin.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003e3.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 19px;\"\u003e\n \u003cp\u003eThe combined incidence of AAA repair or mortality due to AAA rupture\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003eGolledge Jonathan\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e2019\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003eAustralia\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003eCohort study B\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 5px;\"\u003e\n \u003cp\u003e763\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003ePatients with asymptomatic unrepaired AAA of any diameter \u0026ge; 50mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e1. Diabetes with no history or medically recorded；\u003c/p\u003e\n \u003cp\u003e2. Diabetes prescribed metformin at the time of recruitment;\u003c/p\u003e\n \u003cp\u003e3. Diabetes not prescribed metformin at the time of recruitment.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003e3.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 19px;\"\u003e\n \u003cp\u003eThe combined incidence of AAA repair or mortality due to AAA rupture\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003eGolledge Jonathan\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e2017\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003eAustralia and New Zealand\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003eCohort study A\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 5px;\"\u003e\n \u003cp\u003e1357\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003eElderly patients with AAA (infrarenal aortic diameter\u0026ge; 30mm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e1. Diabetes who were prescribed metformin；\u003c/p\u003e\n \u003cp\u003e2. Diabetes not prescribed metformin;\u003c/p\u003e\n \u003cp\u003e3. Patients who neither had diabetes nor were receiving metformin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003e3.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 19px;\"\u003e\n \u003cp\u003eMaximum aortic diameter growth measured by contrast enhanced CT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003eGolledge Jonathan\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e2017\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003eAustralia and New Zealand\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003eCohort study B\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 5px;\"\u003e\n \u003cp\u003e287\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003eElderly patients with AAA (infrarenal aortic diameter\u0026ge; 30mm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e1. Diabetes who were prescribed metformin;\u003c/p\u003e\n \u003cp\u003e2. Diabetes not prescribed metformin;\u003c/p\u003e\n \u003cp\u003e3. Patients who neither had diabetes nor were receiving metformin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003e2.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 19px;\"\u003e\n \u003cp\u003eMaximum aortic diameter growth measured by contrast enhanced CT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003eGolledge Jonathan\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e2017\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003eAustralia and New Zealand\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003eCohort study C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 5px;\"\u003e\n \u003cp\u003e53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003eElderly patients with AAA (infrarenal aortic diameter\u0026ge; 30mm)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e1. Diabetes who were prescribed metformin;\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e2. Diabetes not prescribed metformin;\u003c/p\u003e\n \u003cp\u003e3. Patients who neither had diabetes nor were receiving metformin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 19px;\"\u003e\n \u003cp\u003eMaximum aortic diameter growth measured by contrast enhanced CT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003eItoga Nathan K\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e2019\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003eUnited States\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003eCohort study\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 5px;\"\u003e\n \u003cp\u003e13,843\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003eElderly patients with diabetes and a diagnosis of AAA without rupture\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e1. Diabetes prescribed metformin\u003c/p\u003e\n \u003cp\u003e2. Diabetes not prescribed metformin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003e4.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 19px;\"\u003e\n \u003cp\u003eMaximum aortic diameters growth determined from radiographic reports\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003eUnosson Jon\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e2021\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 9px;\"\u003e\n \u003cp\u003eSewden\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 10px;\"\u003e\n \u003cp\u003eCohort study\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 5px;\"\u003e\n \u003cp\u003e98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 12px;\"\u003e\n \u003cp\u003ePatients with initial abdominal aortic diameter \u0026ge;30mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 18px;\"\u003e\n \u003cp\u003e1. Metformin (T2DM)\u003c/p\u003e\n \u003cp\u003e2. Without metformin (T2DM)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 7px;\"\u003e\n \u003cp\u003e3.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 19px;\"\u003e\n \u003cp\u003eMaximum aortic diameter growth measured by ultrasound\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 4px;\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e3\u0026nbsp;\u003c/strong\u003eMediating effects of TC and LDL\u003c/p\u003e\n\u003cdiv align=\"Left\"\u003e\n \u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"756\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 94px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMediator\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 281px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eOR of \u0026nbsp;Intermediary effect (95% Cl)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eb_mediation\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 92px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ese_mediation\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 174px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMediation efect proportion\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 94px;\"\u003e\n \u003cp\u003eLDL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 281px;\"\u003e\n \u003cp\u003e0.271 (0.133-0.552)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 115px;\"\u003e\n \u003cp\u003e-1.3057\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 92px;\"\u003e\n \u003cp\u003e0.3627\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 174px;\"\u003e\n \u003cp\u003e0.2884\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 94px;\"\u003e\n \u003cp\u003eTC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 281px;\"\u003e\n \u003cp\u003e0.3199 (0.159-0.644)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 115px;\"\u003e\n \u003cp\u003e-1.1397\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 92px;\"\u003e\n \u003cp\u003e0.3572\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 174px;\"\u003e\n \u003cp\u003e0.2517\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Abdominal Aortic Aneurysm, Metformin, Mendelian Randomization, Pharmacotherapy, Therapeutic Targets, Molecular Mechanisms","lastPublishedDoi":"10.21203/rs.3.rs-6834737/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6834737/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eObjective: \u003c/strong\u003eOur research sought to investigate the relationship between metformin and abdominal aortic aneurysm (AAA) risk. We aimed to contribute to the current understanding of metformin's potential as a pharmaceutical intervention for AAA and explore the genetic facets influencing its effect.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003eThe methods encompassed meta-analysis, Mendelian Randomization (MR) and colocalization analysis. Meta-analysis facilitated a robust review and synthesis of the current literature. MR could investigate the causal relationship between metformin treatment and AAA incidence, and the mediating effects of certain factors. Colocalization analysis results are used as a supplement to MR analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eMeta-analysis suggest that metformin prescription is associated with a clinically important significant reduction in both growth in people with AAA. MR analysis results indicated the negative correlation between metformin treatment and AAA (OR = 0.0108, 95% CI: 0.000204 - 0.572, \u003cem\u003eP\u003c/em\u003e = 0.0254), and the reduction of total cholesterol levels (mediation effect 25.17%, OR=0.3199, 95% CI: 0.159 - 0.644) mediated the protective effect of metformin against AAA, and the main role may be played by low-density lipoprotein levels (mediation effect 28.84%, OR = 0.271, 95% CI: 0.133 - 0.552). Colocalization analysis identified 9 significant genes, notably GPD2, associated with both metformin treatment and reduced AAA risk.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions: \u003c/strong\u003eMetformin treatment is associated with a significant reduction in AAA incidence, potentially mediated by its effects on blood lipid levels. These findings support further investigation into metformin as a therapeutic option for AAA, emphasizing the importance of integrating pharmacological and genetic approaches in cardiovascular disease management.\u003c/p\u003e","manuscriptTitle":"Deciphering the Role of Metformin in Abdominal Aortic Aneurysm Progression: Insights from Two-Step Mendelian Randomization and Colocalization Analysis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-27 14:35:31","doi":"10.21203/rs.3.rs-6834737/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"98ba1eae-a02d-4a49-891d-4cf6faea8111","owner":[],"postedDate":"June 27th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-08-01T01:38:28+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-27 14:35:31","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6834737","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6834737","identity":"rs-6834737","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}