Causal Relationship between Gut Microbiota and Pulmonary Embolism: An Analysis Using Mendelian Randomization

Venous thromboembolism (VTE) encompasses pulmonary embolism (PE) and deep vein thrombosis. A sudden onset characterizes PE and carries the potential for fatality, with limited options for early detection and management of changeable risk factors. The pathogenesis of PE is multifaceted, involving a mix of environmental and genetic risk factors, with inflammation having a significant impact on the advancement of this condition (Lutsey and Zakai 2023). Inflammation serves to activate endothelial cells, platelets, and white blood cells, thereby initiating clotting processes and contributing to the development of coagulation disorders as well as increased pro-inflammatory cytokine levels, chemokines, and diverse white cell subtypes (Saghazadeh and Rezaei 2016). The gut microbiome (GM) is connected to the etiology of numerous inflammatory disorders, with intestinal dysbiosis being linked to an increased prevalence of venous thromboembolism (Kappelman et al. 2011; Freuer et al. 2022). The dysregulation of GM can induce the involvement of innate immune cells, platelets, and vascular endothelial cells in inflammatory pathways, which release multiple clotting proteins and promote a pro-thrombotic state (Lutsey and Zakai 2023). Concurrently, recent research has extensively investigated the “gut-lung axis” relationship, revealing additional connections between intestinal flora metabolites and alterations in pulmonary function. Data indicates that the composition of the gut microbiome influences the pathogenesis of long-term pulmonary obstruction (Hasan et al. 2020) and asthma (Barcik et al. 2020) through its regulatory impact on inflammation. While there is a known correlation between intestinal dysbiosis and a heightened risk of VTE, the specific molecular pathways through which gut microbiota contribute to the development of VTE, particularly PE, have not yet been thoroughly investigated.

After acute myocardial infarction and stroke, VTE, which includes PE and deep vein thrombosis, is the third most common vascular disease, with an estimated 10 million cases annually, posing a significant global health challenge (Raskob et al. 2014). PE, a frequently encountered and potentially life-threatening condition in the emergency department, typically results in mortality within a short timeframe post-diagnosis. PE can lead to a range of pathophysiological disorders, with clinical thrombosis resulting from of vascular wall damage, alterations in blood flow, and abnormalities in blood components (Ballard et al. 2012). The development of PE is intricate and is impacted by a confluence of environmental and genetic variables. Increasing data indicates that inflammation contributes significantly to the risk of developing PE (Zhang et al. 2023). Inflammation not only triggers the coagulation process, but also contributes to coagulation dysfunction and an elevation in pro-inflammatory cytokines, chemokines, and different types of white blood cells. Activated white blood cells are the primary origin of tissue factor positive procoagulant microparticles, which have the potential to induce thrombus formation and growth (Zhang et al. 2023). Despite significant progress in identifying and treating lung embolism, further research is warranted to identify modifiable risk factors in a timely manner for effective prevention and treatment. The GM is known to perform a significant role in the pathogenesis of obesity, sepsis/infection, inflammatory bowel disease, and intestinal failure and has been associated with a higher risk of VTE (Miehsler et al. 2004; Ley et al. 2006; Kaplan et al. 2015; Gonzalez-Hernandez et al. 2016). Disruption of the GM by environmental or genetic influences can trigger inflammatory pathways in innate immune cells, platelets, and vascular endothelial cells, which causes coagulation protein to be released and promotes an inflammatory response, leading to a prothrombotic state (Zhu et al. 2016; Jäckel et al. 2017). Research has demonstrated that dysregulation of the GM can result in cellular dysfunction and tissue damage, which are collectively referred to as lesions in multiple organs, including the lungs, a phenomenon known as the gastrointestinal-lung axis (Budden et al. 2017). The gastrointestinal-lung axis is a bidirectional communication system between the gut and the lungs, which is mainly mediated by the immune, nervous, and circulatory systems. In the gut, the normal gut microbiota plays a crucial role in maintaining intestinal homeostasis. When the gut microbiota is dysregulated, such as antibiotic use, diet changes, or infections, it can lead to increased intestinal permeability. This allows endotoxins, such as lipopolysaccharides, and microbial metabolites to translocate from the gut into the bloodstream. Once in the circulation, these substances can reach the lungs via the systemic circulation. In the lungs, they may trigger an inflammatory response, affecting pulmonary function and potentially contributing to the development of lung diseases. Additionally, gut microbiota changes can activate the immune cells in the gut-associated lymphoid tissue. These activated immune cells can then circulate to the lungs through the bloodstream and lymphatic system, modulating the local immune response. Conversely, lung inflammation can also impact the gut microbiota. For instance, cytokines and other inflammatory mediators produced in the lungs during a respiratory infection can reach the gut through circulation, altering the composition and function of the gut microbiota. As such, this axis helps hormones, cytokines, endotoxins, and microbial metabolites enter the bloodstream (Dang and Marsland 2019). These results offer insight into the association between gut microbiota and thromboembolism, suggesting a potential link to PE. Nevertheless, it is unclear how gut microbiota contributes to PE pathogenesis, and this mechanism remains incompletely understood.

The Mendelian randomization (MR) approach utilizes integrated genome-wide association studies (GWAS) data to identify suitable single nucleotide polymorphisms (SNPs) serving as useful variables (IV) in evaluating the causal association among an exposure combined with a result (Birney 2022). MR employs random allocation in allelic inheritance, mitigates the impact of confounding variables, and does not alter the genetic sequence of individuals, facilitating the investigation of the cause-and-effect link between GM and PE via MR research. MR methods have been utilized to evaluate potential causal associations between the gut microbiota (GMS) and lung conditions such as chronic obstructive pulmonary disease, asthma, and thromboembolism (Hasan et al. 2020; Freuer et al. 2022; Wei et al. 2023). However, while several recent studies have explored the relationship between gut microbiota and venous thromboembolism (Wang et al. 2023; Meng et al. 2024; Xi et al. 2024), the specific causal relationship between gut microbiota and PE using MR analysis warrants further investigation.

This research chose GM as the independent variable and PE as the dependent variable in a Mendelian randomization analysis. The study assessed the possible reason for the association between GM and PE through the lens of host genetics, offering a foundational framework for future investigations into the intricate mechanisms underlying PE. Furthermore, the identification of specific GM profiles in PE patients may lead to the finding of new biomarkers as well as the growth of improved diagnostic and therapeutic approaches.

Recently, several MR studies (Wang et al. 2023; Meng et al. 2024; Xi et al. 2024) have indeed investigated the relationship between gut microbiota and VTE. Similar to these studies, our research also found that Slackia and Bacteroides were associated with a reduced risk of VTE. However, our study offers unique contributions. Unlike previous studies limited to specific population subsets or particular gut microbiota, we explored the associations between multiple gut microbiota genera and pulmonary embolism (PE), a specific type of VTE, in a broader context. This broader scope provides a more comprehensive view of the role of gut microbiota in VTE pathogenesis. Regarding research methods, there are differences in the selection of instrumental variables and the emphasis on analysis methods These previous studies may have used different SNP screening criteria or relied more on specific MR methods. Our study, however, comprehensively applied multiple MR methods and conducted a more thorough sensitivity analysis to ensure the reliability of the results.

This research investigated the potential causal link between GM and PE through a genetic perspective utilizing MR techniques. The study employed independent GM-related SNPs identified in GWAS as instrumental variables and evaluated their impact on PE risk using MR methodologies. Following the implementation of the IVW technique as the principal causal analysis, the results indicated that four specific GM taxa - Slackia, Oscillospira, Bacteroides, and Clostridium sensu stricto 1 – correlated with a reduced likelihood of developing PE. A test of pleiotropy and heterogeneity confirmed that these findings were reliable, as no significant pleiotropy or heterogeneity was detected.

Our findings align with some recent studies that have explored the relationship between gut microbiota and pulmonary diseases. For example, similar to our results on Oscillospira, previous research has shown that an increase in SCFA-producing bacteria, such as Oscillospira, is often associated with a reduced inflammatory state in the body, which may contribute to a lower risk of PE. However, it should be noted that in some other studies on lung dysfunction, the role of bacteria can be complex. Some bacteria may have a dual-edged effect, being beneficial in specific contexts but harmful in others. In our study, the inverse relationship between Bacteroides abundance and PE formation contradicts the general pattern in some literature where Bacteroides is sometimes considered a pathogen in the development of various diseases. This discrepancy might be due to differences in study populations, research methods, or the specific subsets of Bacteroides species studied. For instance, the mouse model in Zhang et al.’s study (2023) may not fully represent the human situation, and the diversity of Bacteroides species and their functions in different hosts could lead to such conflicting results.

Regarding new clinical findings, while previous studies mainly focused on the overall relationship between gut microbiota and VTE, our study specifically identified four gut microbiota genera with potential protective effects against PE. This provides more targeted information for future clinical research. For example, it may guide the development of personalized probiotic therapies. Supplementing specific bacteria such as Slackia, Oscillospira, Bacteroides, or Clostridium sensu stricto 1 can modulate the gut microbiota and potentially reduce the risk of PE. This finding also enriches our understanding of the role of gut microbiota in developing PE and offers new ideas for preventing and treating this disease.

This study represents the initial Mendelian randomized investigation into the causal involvement of GM in PE. Employing MR Analysis, we examined the GM taxa and possible causal relationships PE through the lens of host genetics, confirming its impact on altering susceptibility to PE. The composition of the gut microbiome is known to change during disease progression, and in the case of pulmonary embolism, it likely undergoes multiple changes as the disease develops through its earlier stages. The protective bacteria identified in our study, such as Slackia, Oscillospira, Bacteroides, and Clostridium sensu stricto 1, represent a long-term attempt by the microbiome to mitigate adverse responses within the host. This could be achieved through various mechanisms. For example, these bacteria might modulate the immune response by reducing inflammation or enhancing the host’s defense mechanisms. They could also influence the production of metabolites that benefit the host’s physiological state, potentially counteracting the processes that lead to PE. However, further research is needed to confirm this hypothesis. Understanding these potential regulatory mechanisms of the gut microbiome could provide new insights into the pathogenesis of PE and open up new avenues for therapeutic interventions. The protective effects of four genus levels of intestinal bacteria, namely Slackia, Oscillospira, Bacteroides, and Clostridium sensu stricto 1, on PE may offer potential biomarkers for future research. Additionally, our findings suggest the possibility of targeted dysregulation of specific gut microbiota taxa as a method for treating and preventing PE. The development of pulmonary embolism is a complex process that likely involves multiple circulatory and tissue disorders and is not solely due to the direct action of the microbiome or its metabolites. Numerous potential confounding factors could influence the relationship between gut microbiota and PE. General health status plays a crucial role. For example, individuals with pre-existing chronic diseases may have a disrupted gut microbiota composition. At the same time, these diseases can also increase the risk of PE through various pathways, such as inflammation and coagulation disorders. Nutritional status also cannot be ignored. Malnutrition or an imbalanced diet may lead to changes in the gut microbiota, and specific nutrients (or lack thereof) can affect the body’s inflammatory response and blood coagulation function, both related to PE development. Dietary habits, like a diet high in fat or low in fiber, can alter the community structure and function of the gut microbiota. A high-fat diet may promote the growth of certain bacteria associated with an increased risk of inflammation. In contrast, a low-fiber diet may reduce the production of beneficial microbial metabolites such as short-chain fatty acids, which are important for maintaining gut and overall health. These factors interact with the gut microbiota in complex ways and can significantly impact the susceptibility to PE.

Oscillospira is a significant contributor to the synthesis of fatty acids with short chains (SCFA), which serve as a crucial criterion for the evaluation of potential “Probiotics of the next generation” (Yang et al. 2021). SCFA is primarily generated via the fermentation of carbohydrates by intestinal microorganisms and is essential for controlling metabolic and immune equilibrium. Butyric acid, a product of SCFA, serves as a key component in the preservation of intestinal health and has been demonstrated to bolster the function of the intestinal barrier through the upregulation of proteins in tight junctions such as Claudin-1 and Zonula occludens 1 (Wang et al. 2012). Numerous investigations have demonstrated a noteworthy correlation between inflammation and Oscillospira, with most findings indicating a negative correlation (Konikoff and Gophna 2016), aligning with our own research. Xu et al. (2021) observed a negative correlation between the profusion of Oscillospira and disease severity in ulcerative colitis patients, while Lima and Longman (2021) reported a decreased presence of Oscillospira in children’s inflammatory bowel disease intestines. Nevertheless, the precise mechanism by which Oscillospira influences inflammation, including PE, remains uncertain.

Bacteroides, a commensal microorganism in the gut, can become opportunistic when extraintestinal translocation occurs (Zafar and Saier 2021), leading to the development of various disease conditions upon its migration into sterile tissues (Koropatkin et al. 2012). Factors contributing to this translocation include impaired immune function, intestinal barrier disruption, surgical trauma, excessive antibiotic use, and aging (Vaishnavi 2013; Archambaud et al. 2019). Zhang et al. (2023) conducted a study in which a PE mouse model was established using vancomycin to modulate the gut microbiome. The findings indicated that PE mice with reduced levels of Bacteroides exhibited a higher survival rate (Zhang et al. 2023). Conversely, our research revealed an inverse relationship between Bacteroides abundance and PE formation. This paradoxical outcome may be attributed to interspecies variations in our respective studies. Recent research has also shown that Bacteroides can have different effects depending on the stage of development and the overall gut microbiota composition. For example, in the context of infancy, a Bacteroides-dominant gut microbiota may enhance neurodevelopment, while in other situations, it could be involved in disease processes. This further emphasizes the need for a more in-depth understanding of Bacteroides species and their functions in different host-microbe interactions. Nevertheless, it is worth noting that not all instances of extraneous translocations result in adverse health conditions. Previous studies have indicated the significance of Bacteroides as a primary producer of vitamin K, a crucial nutrient synthesized by human gut bacteria (Walther et al. 2013). It can boost bone density to prevent or treat osteoporosis (Fujita et al. 2012). Tamana and colleagues (2021) discovered that in the later stages of infancy development, a gut microbiota dominated by Bacteroides can potentially enhance neurodevelopment. These findings indicate that some Bacteroides species could exhibit both pathogenic and helpful effects, contingent upon their host location. Therefore, a comprehensive understanding of the species and functions of Bacteroides is essential in elucidating their effects on illness and host health within the context of gut microbiome research.

The bacterial genera Slackia and Clostridium sensu stricto 1) have been underrepresented in scientific literature, resulting in a limited understanding of their physiological functions and impact on host health. The study of these lesser-known genera remains a prominent focus in microbiology and medical research, with the potential for significant advancements in our knowledge of their ecological roles and health implications. Further research holds promise for a more comprehensive understanding of Slackia and Clostridium sensu stricto 1 bacteria and their relevance to human health.

The research presents several strengths alongside certain constraints. These strengths include the novel utilization of two samples for MR analysis, allowing for an analysis of potential resonal relationships between GM and PE at the genus level. Additionally, as compared to previous randomized controlled studies, the gut microbiota data we used is the GWAS to date, giving the study a larger sample size. Lastly, the significant epidemiological implications of MR analysis suggest its ongoing relevance and potential expansion later on. It is important to recognize this study’s limitations. Besides the issues we have already mentioned, other potential confounding factors could influence the results. For example, diet is closely related to the composition of gut microbiota. Different dietary patterns, such as high-fiber or high-fat diets, may affect the abundance and function of the gut microbiota, and at the same time, diet can also impact the body’s inflammatory state and coagulation function, which are relevant to the development of PE. Lifestyle factors like exercise and smoking also cannot be ignored. Regular exercise may modulate the gut-lung axis through various mechanisms, and smoking can disrupt the balance of the gut microbiota and increase the risk of inflammation and thrombosis. However, due to the nature of our data source, these factors were not accounted for in the current analysis. In addition, it is important to recognize this study’s limitations, including the following: i) While our study supports the MR hypothesis, it does not definitively eliminate the potential for weak instrument bias. ii) Incorporating individuals of European descent in the GWAS data may limit the study’s findings’ applicability to other ethnic populations. iii) The exposure data only classified genus, restricting the ability to analyze relationships at the species level. Insufficient IV availability for reverse MR analysis precludes the determination of the possible link of causation between GM and PE. To address the concern about the generalizability of our findings, we acquired GWAS data of PE from https://ngdc.cncb.ac.cn/omix/release/OMIX001381 and GWAS data of gut microbiota from https://db.cngb.org/search/project/CNP0000794 for Asian populations. Repeating the Mendelian randomization analysis with these Asian datasets, we found that some of the gut microbiota genera, such as Slackia, Oscillospira, Bacteroides, and Clostridium sensu stricto 1, still showed significant associations with the risk of PE in the Asian population. However, the strength of these associations differed from those observed in the European population used in our original study. These differences may be attributed to genetic, environmental, and lifestyle variations among ethnic groups. For example, genetic background can influence the composition and function of the gut microbiota, and dietary habits, which vary across ethnicities, can also play a role in modulating the gut-lung axis. This finding emphasizes the importance of researching diverse ethnic populations to understand the relationship between gut microbiota and PE fully.

While bacteria are the major component of the gut microbiome, other components such as the mycobiome, viruses, archaea, and protists can also significantly modulate circulatory and tissue dysfunction. For example, certain fungi in the gut mycobiome may interact with the immune system and influence inflammation, which is closely related to the development of PE. Viruses in the gut virome can affect the gut epithelial barrier function, potentially leading to changes in the translocation of microbial products and subsequent effects on the host’s health. Archaea may contribute to the metabolism of the gut microbiota, and protists can interact with bacteria and the host immune system in complex ways. However, in the present study, we mainly focused on bacteria and did not fully consider the impact of these other microbial components. This is a limitation of our study. Future research should aim to comprehensively investigate the relationships between all components of the gut microbiome and PE to gain a more complete understanding of the underlying mechanisms.

Hence, this finding represents a preliminary investigation at the species level. Subsequent research endeavors intend to broaden the sample size and investigate the correlation between GM taxa and PE at the species level to offer enhanced theoretical backing for investigating the “gut-lung” axis mechanism.

Our study employs a Mendelian randomization analysis with two samples using publicly available GWAS aggregate research to establish the causal relationship among GM (specifically Slackia, Oscillospira, Bacteroides, and Clostridium sensu stricto 1) and PE. The outcomes of our investigation suggest that gut flora could significantly influence the onset and progress of PE, offering a novel understanding of the processes through which gut microbiota impacts the susceptibility to lung diseases, specifically PE.

These findings have important implications for future research and clinical practice. Regarding potential therapeutic interventions, we could consider developing probiotic-based therapies. For example, supplementing with Slackia, Oscillospira, Bacteroides, or Clostridium sensu stricto 1 strains may help modulate the gut microbiota and potentially reduce the risk of PE. Additionally, dietary interventions that promote the growth of these beneficial bacteria could be explored. This might involve increasing the intake of prebiotic foods rich in fiber, which can serve as substrates for the growth of these bacteria.

On the clinical side, the identified associations could be used to develop new risk prediction models for PE. By incorporating information about an individual’s gut microbiota composition, we can more accurately assess their risk of developing PE, especially in high-risk populations. This could enable earlier preventive measures, such as targeted anticoagulant therapy or lifestyle modifications.

Furthermore, these discoveries could offer a theoretical framework for advancing innovative interventions and therapies for PE targeted at enhancing or altering GM.