Introduction

N-terminal pro-B-type natriuretic peptide (NT-proBNP), a derivative of the precursor molecule of B-type natriuretic peptide (BNP), serves as a pivotal biomarker for evaluating cardiovascular function and assessing the risk of cardiac disease, particularly heart failure¹ Li-natriuretic peptide-guided therapy improves cardiovascular outcomes of patients with high-risk heart failure and reduced ejection fraction disease.² In-hospital 24-hour blood pressure monitoring in patients with decompensated atrial fibrillation (AF) revealed that patients with a dipper blood pressure pattern exhibited a favorable association with the change of NT-proBNP levels.³ Patients with coronary slow flow phenomenon with scar tissue on cardiac magnetic resonance imaging have higher NT-proBNP levels.⁴ In addition, several studies have noted that NT-proBNP contributes to the incidence of coronary artery disease by increasing the risk of heart failure disease and is an important assessment indicator of coronary artery disease.^5-7 Despite the fact that diseases caused by heart failure are treated clinically with appropriate measures, including medications, interventions, or cardiac surgery, heart failure continues to threaten humans as a common condition. Therefore, modifying NT-proBNP levels is crucial for preventing and treating heart failure.

The collective microbial community of all microorganisms in the human gut is termed “gut microbiota (GM).”⁶ The host provides the environment and nutrients necessary for GM survival, and GM actively participates in various physiological processes within the human body in turn. In recent years, studies have found a close relationship between GM and heart failure. Some researchers have detected identical bacterial DNA in arterial plaques and the intestinal flora of coronary heart disease (CHD) patients, suggesting that GM may be a potential source of bacteria in arterial plaques, and it may be involved in the development of CHD.⁷ Although heart failure contributes to cardiovascular diseases such as CHD⁸ and myocardial infarction,⁹ and GM is associated with heart failure-induced cardiovascular disease,¹⁰ the causality of GM and heart failure remains debated. Modulating or remodeling the GM promises to be a new clinical therapeutic idea for the effective and specific treatment of heart failure.

Mendelian randomization is a statistical method that emphasizes overcoming confounders and reverse causality bias. This approach is based on the fact that IVs are associated with the risk factor or biomarker being studied but not with other potential confounders. Thus, these instrumental variables (IVs) can be considered randomly assigned to individuals, similar to random assignment in an experiment. By adopting this approach, causality can be assessed more reliably without interference from other factors, increasing the credibility of study findings.¹¹ Although randomized controlled trials (RCTs) are the gold standard for evaluating the safety and efficacy of an intervention, they are challenging to implement because of the complexity of their design, the high cost in terms of time, human and material resources, and ethical constraints. Conceptually, MR is similar to RCTs in that genetic variables are randomly assigned at birth to “case” or “control” groups as IVs and remain constant throughout the life cycle, according to Mendel’s second law. Genome-wide association study (GWAS) has played an essential role in complex disease genetics by searching for genotype-phenotype relationships through millions of genetic variations tested on the genomes of many individuals.¹² Single nucleotide polymorphism (SNP) exposure and SNP-outcome associations can be extracted separately from independent GWAS and used in MR analyses of 2 independent samples to generate a single causal estimate. Thus, these enabled us to infer a causal relationship between GM and NT-proBNP using a Mendelian randomization (MR) approach. It is independent of confounders between genetic variation and outcome, and serves as a novel approach to studying causal relationships between GM and NT-proBNP.^13-15 Therefore, the relationship between GM and NT-proBNP was investigated from a Mendelian genetic perspective.

Methods

Study Design

Intestinal flora and NT-proBNP were respectively used as the exposure factor and outcome variable, with significantly associated SNPs serving as IVs. Based on publicly available GWAS summary statistics, the causal relationship between 119 bacterial taxa and NT-proBNP was assessed using 2-sample bidirectional MR analysis. This study adhered to 3 crucial assumptions of the MR methodology: (1) there is a significant association between IVs and gut flora; (2) IVs are not associated with any of the gut flora-NT-proBNP confounders; and (3) IVs affect NT-proBNP only through the gut flora, instead of any other pathway.

Data Source

Genetic data on gut flora were obtained from the GWAS meta-analysis published by the largest MiBioGen consortium, which collected 16S rRNA gene sequencing profiles and genotyping data from 18 340 subjects (13 266 of European origin) from 11 countries, spanning Europe and the United States, and analyzed the gut microbiome signature loci.¹⁶ In this study, bacterial taxa were analyzed at the genus level, encompassing 119 bacterial taxa after excluding 15, and a total of 3301 NT-proBNP sources, all of European origin,¹⁷ were included. The NT-proBNP was from a population with a rigorously collected set of demographic data, anthropometric measurements, lifestyle factors, and dietary information. To ensure a clean baseline for outcome assessment, participants with a history of major diseases, recent illnesses, or infections were excluded from the study. The NT-proBNP levels were measured using SomaLogic assays, which employ SOMAmer probes to accurately identify and bind target proteins. This advanced method allows for the detection of proteins even at low concentrations, providing precise and reliable measurements essential for evaluating NT-proBNP levels. The NT-proBNP datasets were chosen primarily from European ancestry populations, consistent with the fact that the GM exposure data, which also predominantly comes from European ancestry. This was done to maintain consistency and reduce confounding variables related to genetic and environmental diversity.

Instrumental Variable

First, SNPs with significant association with GM (P < 1 × 10⁻⁵) were selected. Second, SNPs were quality checked according to the following quality control steps to ensure accuracy in establishing a causal relationship between GM and NT-proBNP: (1) exclusion of SNPs with allelic incongruence between the exposure and outcome samples, such as A/C; (2) removal of palindromic SNPs; (3) setting a coefficient r ² threshold of 0.001 and a region width of 10 000 kb to exclude the interference of linkage disequilibrium; (4) remove minor allele frequency. The MR-Egger regression was used to assess the average level of pleiotropy of the SNPs as a whole, whereas MR-PRESSO could detect each abnormal SNP and re-analyze the MR analysis by removing the SNPs with level of pleiotropy. For significantly associated SNPs, the F-statistic was used to assess whether the selected genetic variables were weakly instrumental. An F-value greater than 10 indicated that the genetic variables did not have weak instrumental bias.

Mendelian Randomization Analysis

In this study, 5 methods were used to verify the causal relationship between GM and NT-proBNP in 119 taxa: MR-random or fixed effects inverse variance weighted (IVW), weighted median estimation (WME), MR-Egger regression, simple mode (SM), and weighted mode (WM). For features containing only 1 IV, the Wald ratio test was used to estimate the association between the identified IV and NT-proBNP. Results with multiple IVs were mainly based on the IVW method, supplemented by the other 4 methods.

Sensitivity Analysis

To ensure the reliability of the causal effect assessment results, sensitivity analysis was conducted in this study. The leave-one-out tests were performed to examine the effect of individual IVs on the overall results. A test for heterogeneity was performed using Cochran’s Q statistics to assess whether there were differences between IVs in the 2-sample MR analysis. Q statistics significant at a P-value < .05 were considered heterogeneous. Horizontal pleiotropy is indicated if the IV can directly affect the outcome variable without exposure, which violates the exclusivity assumption of MR. This study used the MR-Egger intercept test to examine horizontal pleiotropy. An intercept term approaching 0 with P > .05 indicated no horizontal pleiotropy; otherwise, horizontal pleiotropy was present. Additionally, the MR-PRESSO was used as a supplementary method to detect horizontal pleiotropy.

Statistical Analysis

Data were organized and analyzed using R software (version 4.2.0). The “TwosampleMR” and “MR-PRESSO” R packages were utilized for MR analysis. Evaluate the causal relationship between GM and NT proBNP using IVW, WME, MR-Egger regression, SM, and WM algorithms. Cochran’s Q statistical tests were conducted to confirm heterogeneity among the selected IVs, and horizontal gene pleiotropy tests were performed on the MR-Egger algorithm. Perform sensitivity analysis on IV SNPs using the leave-one-out test. Statistical significance was set at P < .05.

Results

Instrumental Variables Selection

Based on the IVW method, a total of 9 out of 104 GM classes were found to be causally associated with NT-proBNP (P < .05). These included 13 SNPs in the genus Holdemanella, 9 SNPs in the genus Coprococcus 2, 11 SNPs in the genus Ruminococcaceae UCG 004, 8 SNPs in the genus, 8 SNPs in the genus Actinomyces, 15 SNPs in the genus Alistipes, 12 SNPs in the genus Lachnospiraceae UCG 008, 9 SNPs in the genus Eubacterium fissicatena group, 12 SNPs in the genus Eubacterium rectale group, and 15 SNPs in the genus Eubacterium ventriosum group. All F values were greater than 10, indicating a low likelihood of a weak IV bias, consistent with the MR association hypothesis.

Mendelian Randomization Analysis of Gut Microbiota and N-Terminal Pro-B-Type Natriuretic Peptide

As shown in Table 1, the genus Holdemanella (β = −0.19, 95% CI: −0.36 to −0.01, P = .037), genus Coprococcus 2 (β = −0.27, 95% CI: −0.54 to 0.00, P = .047), genus Ruminococcaceae UCG 004 (β = -0.23, 95% CI: −0.45 to −0.02, P = .032), genus Alistipes (β = −0.29, 95% CI: −0.55 to −0.03, P = .031) were negatively associated with NT-proBNP. Conversely, the genus Actinomyces (β = 0.22, 95% CI: 0.01-0.44, P = .042), genus Lachnospiraceae UCG 008 (β = 0.27, 95% CI: 0.1-0.44, P = .002), genus Eubacterium fissicatena group (β = 0.17, 95% CI: 0.01-0.32, P = .033), genus Eubacterium rectale group (β = 0.4, 95% CI: 0.13-0.67, P = .003), and genus Eubacterium ventriosum group (β = 0.26, 95% CI: 0.02-0.49, P = .032) were positively associated with NT-proBNP.

Consistent with the above results, the overall direction of causality between GM and NT-proBNP calculated by the 5 methods is shown in Figures 1 and 2. Among them, the genus Holdemanella, genus Coprococcus 2, genus Ruminococcaceae UCG 004, and genus Alistipes appear to exert reduced effects on NT-proBNP. While the genus Actinomyces, genus Lachnospiraceae UCG 008, genus Eubacterium fissicatena group, genus Eubacterium rectale group, and genus Eubacterium ventriosum group appear to act as increased factors for proBNP. The influence and range of fluctuation of each SNP in the causal relationship between GM (exposure factor) and NT-proBNP (outcome factor) is illustrated in the figures. The contribution of each SNP to the overall combined effect value of GM (exposure factor) and NT-proBNP (outcome factor) causality is shown in Figure 2.

Sensitivity Analysis Results

The test for heterogeneity of the gut microorganisms of the 9 taxa showed P > .05, indicating no heterogeneity. The intercept term of MR-Egger’s test was statistically non-significant (P > .05), and therefore the assumption of exclusivity was considered to be valid, as well as the causal relationship was robust. As shown in funnel plot (Figure 3), the points represented by individual SNPs are roughly symmetrically distributed, indicating a low likelihood of being affected by potential bias. The results of the leave-one-out method show that there is no big difference between the effect sizes of individual SNPs before and after the exclusion, suggesting that there is no individual SNP that significantly affects the MR estimation results, which indicates that the causal relationship has a certain degree of stability, as shown in Figure 4.

Discussion

Based on MR analysis, a causal association was found between 9 intestinal bacterial species and NT-proBNP. Among them, genera Holdemanella, Coprococcus 2, Ruminococcaceae UCG 004, and Alistipes were reduced factors for NT-proBNP; while genera Actinomyces, Lachnospiraceae UCG 008, Eubacterium fissicatena group, Eubacterium rectale group, and Eubacterium ventriosum group were increased factors for NT-proBNP. Several studies have noted that increased NT-proBNP promotes heart failure, which further exacerbates cardiovascular disease.^18,19 Therefore, regulating the GM levels mentioned above is beneficial in improving outcomes of cardiovascular diseases such as heart failure caused by NT-proBNP.

It is known that BNP is expressed by the myocardium in response to elevated atrial wall pressure, both BNP as well as NT-proBNP have become important biomarkers of cardiovascular disease.²⁰ Quantification of NT-proBNP has been shown to predict the progression of coronary artery disease. One previous study had found different doses of azithromycin affected the abundance and grouping of GM in rat models of heart failure, and the degree of myocardial damage, GM variations, and clinical manifestations of heart failure were found to be consistent with each other.²¹ Another study noted that hypertension could result in the imbalance of GM, while chronic hypertension could lead to dysregulation of fecal microbial ecology and elevated levels of the biomarker NT-proBNP.²² Although the above studies indicate that gut flora may be associated with NT-proBNP-induced heart failure disease, no studies have directly demonstrated a causal relationship between GM and NT-proBNP.

In the present study, it was found that the genus Holdemanella, genus Coprococcus 2, genus Ruminococcaceae UCG 004, and genus Alistipes were reduced factors for NT-proBNP. Consistent with our findings, Wang et al.²² found that the structure of several GM, including Holdemanella, was associated with response to statins. Therefore the authors hypothesized that manipulating the composition of the GM may be a clinical treatment for controlling lipids in patients with coronary artery disease . However, no clinical studies have shown that regulating the genus Holdemanella can alleviate diseases such as heart failure caused by NT-proBNP. Reducing Alistipes promoted lipid reduction and delayed ventricular remodeling after myocardial infarction.²³ A randomized controlled study found that differential gut microbiome profiling could be used to identify patients with AF and that reduced Alistipes abundance was detected in patients with AF.²⁴ Additionally, the genus Coprococcus 2 was identified as a protective factor against atherosclerosis in another MR study,²⁵ and the genus Ruminococcaceae UCG 004 was also found to be a protective factor for AF flutter.²⁶ In summary, both GM and NT-proBNP are closely associated with cardiovascular diseases, and the genus Holdemanella, genus Coprococcus 2, genus Ruminococcaceae UCG 004, as well as genus Alistipes have been identified as increased factors linked to NT-proBNP. Therefore, modulating these genera of bacteria may help alleviate NT-proBNP-induced cardiovascular diseases. The GM can be regulated by diet and fecal transplants. Although the above GM has not been used clinically, the findings provide a research basis for preventing diseases such as heart failure by intervening in GM.

It was found that the genera, UCG 008, group, group, and group were increased factors for NT-proBNP. The genera Actinomycetes and Eubacterium rectale group have also been associated with coronary artery disease as well as adverse cardiovascular events.^27,28 The genera Actinomyces was orally colonized in mice and was beneficial in alleviating pneumonia in mice.²⁹ The results illustrate that the hearts and central nervous systems of the colonized mice did not show any abnormalities, and it is speculated that colonizing genera may be a feasible strategy for preventing and controlling heart failure and other diseases. One previous study observed significantly higher levels of Eubacterium rectale group in the gut flora of individuals in hypertensive group compared to the control group.³⁰ Eubacterium ventriosum group and NT-proBNP were found to be correlated and associated with myocardial infarction in a RCT.³¹ Although no clinical trials have confirmed definitive associations between the aforementioned GM and NT-proBNP, based on the previous evidence, it can be hypothesized that an imbalance in GM may lead to NT-proBNP abnormalities, further exacebating cardiovascular diseases.

The NT-proBNP is an essential biomarker for diagnosing heart failure in clinical practice. The level of NT-proBNP increases with the severity of heart failure, and chronic heart failure patients often have intestinal dysfunction.³² The GM can regulate and maintain intestinal and overall host health by releasing secretions. However, there is currently no direct evidence to suggest the mechanism by which GM regulates NT pro-BNP in the research findings. Studies have shown that GM products short-chain fatty acid (SCFA) can promote intestinal fermentation processes, which is beneficial for the prognosis of heart failure patients.³³ Whereas in the results, genus Holdemanella, genus Coprococcus 2, genus Ruminococcaceae UCG 004, genus Actinomyces, genus Lachnospiraceae UCG 008, genus Eubacterium fissicatena group, genus Eubacterium rectale group, and genus Eubacterium ventriosum group all produce secreted SCFA. Therefore, it is hypothesized that these GMs can regulate the intestinal microenvironment through the production of secreted SCFA, thus further regulating NT-proBNP-induced heart failure and other diseases. In addition, Actinomyces metabolites include vitamin B12 and lactic acid. Patients with heart failure are often deficient in vitamin B12.³⁴ Arterial blood lactate concentration is positively correlated with NT-proBNP in patients with acute heart failure and both levels are abnormal³⁵ It is hypothesized that Actinomyces can regulate diseases such as heart failure caused by NT-proBNP through multiple pathways. The metabolites of Eubacterium fissicatena group also include secondary bile, and the primary and secondary bile acid pathways of GM are involved in cardiovascular and cerebrovascular diseases.³⁶ Elevated levels of NT-proBNP in people without manifestations of heart failure suggest a potential risk of cardiovascular diseases, such as CHD, hypertension, etc. They can be used for early screening and cardiovascular disease risk stratification.³⁷ The SCFA are involved in human blood pressure homeostasis.³⁸ In addition, specific bile acids regulate host lipid metabolism and glucose/insulin metabolism.³⁹ Therefore, it is hypothesized that GM regulates cardiovascular diseases caused by NT-proBNP. Although no study has shown the mechanism of action of the 9 GMs and NT-proBNP, it is speculated that these GMs may regulate the level of NT-proBNP through their own metabolites, which may be further involved in the progression of heart failure and other diseases. Of course, the above is only a simple speculation of the mechanism, and the specific mechanism of action needs to be further verified by experiments.

Although it has been suggested that the GM would be a potential therapeutic target for heart failure as it plays a key role in regulating host physiology and metabolism and in the development of heart failure. However, the discovery of 9 GMs with a causal relationship with heart failure has not been applied in the clinical setting. Only studies have used genera Actinomyces to mitigate disease progression by oral colonization in mice.²⁹ Dietary interventions, probiotic therapy, fecal microbiota transplantation, and antibiotics are all ways to regulate GM. As research continues, the prevention and treatment of disease through dietary modification and oral colonization will be one of the directions for clinical disease treatment.⁴⁰

This study used large-scale GWAS data and MR methods to analyze the causal relationship between 104 GM classes and NT-proBNP. The present study, due to its large sample size, helped to reduce sampling error and was able to help exclude the effect of confounding factors, resulting in a more accurate assessment of the causal relationship between GM and NT-proBNP. In addition, the present study conducted several sensitivity analyses to provide comprehensive quality control on aspects such as multiplicity and heterogeneity of MR analyses, and thus, the conclusions obtained are robust.

The present study provides the most comprehensive assessment of the causal relationship between GM and NT-proBNP based on MR methods, but this study has some limitations. First, due to the limited current understanding of the etiology and mechanisms of abnormal NT-proBNP levels in humans, if genetic tools influence NT-proBNP through confounding variables other than exposure variables, it may introduce observational bias, leading to inaccurate causal estimates. Based on these potential confounding variables, an inherent limitation of the MR method, the results will also be validated using RCTs in the future. Second, there may be gene-environment interactions in the effect of SNPs on exposure factors, implying that SNPs may have a nonlinear impact on the risk of NT-proBNP abnormalities; however, MR analyses can only assess linear associations and cannot evaluate the effect of extremes of exposure factors on disease. In addition, the results of the MR analyses in this study may not be generalizable to populations outside of European ancestry or populations residing in different geographic regions, as genetic heterogeneity and environmental factors vary between areas. This data was not analyzed for specific stratification by gender, age, dietary habits, etc. Finally, complex physiological mechanisms are involved in causing NT-proBNP abnormalities, and the MR analyses performed in this study may not fully capture this causal relationship. Therefore, if GM is to be regulated as a clinical measure to combat diseases such as heart failure caused by NT-proBNP abnormalities, it must be refined by completing higher-quality MR analyses or RCTs. In addition, this study could not point out the specific mechanisms by which GM exerts its effects on NT-proBNP, and further studies are needed to reveal the pathophysiological mechanisms underlying these causal relationships.

Conclusion

In summary, this study explored the causal relationship between GM and NT-proBNP applying MR analysis. Of note, this study indicates that regulating gut flora may potentially prevent and control NT-proBNP abnormalities, thereby reducing the risk of NT-proBNP-related heart failure and other diseases, which lays the foundation for further exploring the mechanism linking GM with heart failure and other diseases. It is also acknowledged that the study’s focus on populations of European ancestry might limit the generalizability of the findings to other populations.

Footnotes

Ethics Committee Approval: This study was approved by the Ethics Committee of Tangshan Fengnan District Hospital (Approval no.: 20240426; Date: April 26, 2024).

Informed Consent: Informed consent was obtained from all individual participants included in the study.

Peer-review: Externally peer-reviewed.

Author Contributions: Concept – Y.W., J.Y.; Design – Y.W., J.Y.; Supervision – J.Y.; Resources – Y.W., J.Y., J.Zheng., J.Zhao., Y.Y., D.G., W.Q., Y.L.; Materials – Y.W., J.Y., J.Zheng., J.Zhao., Y.Y., D.G., W.Q., Y.L.; Data Collection and/or Processing – Y.W., J.Zheng., J.Zhao., Y.Y., D.G., W.Q., Y.L.; Analysis and/or Interpretation – Y.W., J.Y., J.Zheng., J.Zhao., Y.Y., D.G., W.Q., Y.L.; Literature Search – Y.W., J.Y.; Writing – Y.W., J.Y., J.Zheng.; Critical Review – J.Y.

Declaration of Interests: The authors have no conflicts of interest to declare.

References

1. Averina M, Stylidis M, Brox J, Schirmer H. NT-ProBNP and high-sensitivity troponin T as screening tests for subclinical chronic heart failure in a general population. ESC Heart Fail. 2022;9(3):1954-1962.
2. Felker GM, Anstrom KJ, Adams KF. Effect of natriuretic peptide-guided therapy on hospitalization or cardiovascular mortality in high-risk patients with heart failure and reduced ejection fraction: a randomized clinical trial. JAMA. 2017;318(8):713-720.
3. Candemir M, Kızıltunç E, Nurkoç SG, Cihan B, Şahinarslan A. Predictors of length of hospital stay and in-hospital adverse events in patients with acute decompensated heart failure: in-hospital 24-hour blood pressure monitoring data. Hellenic J Cardiol. ;():-.
4. Candemir M, Şahinarslan A, Yazol M, Öner YA, Boyacı B. Determination of myocardial scar tissue in coronary slow flow phenomenon and the relationship between amount of scar tissue and Nt-ProBNP. Determinação do Tecido Cicatricial do Miocárdio no Fenômeno de Fluxo Coronário Lento e a Relação entre a Quantidade de Tecido Cicatricial e o Nt-ProBNP. Arq Bras Cardiol. 2020;114(3):540-551.
5. Sterling MR, Durant RW, Bryan J. N-terminal pro-B-type natriuretic peptide and microsize myocardial infarction risk in the reasons for geographic and racial differences in stroke study. BMC Cardiovasc Disord. 2018;18(1):66-.
6. Vemuri R, Gundamaraju R, Shastri MD. Gut microbial changes, interactions, and their implications on human lifecycle: an ageing perspective. BioMed Res Int. 2018;2018():4178607-.
7. Guo Y, Li H, Liu Z. Impaired intestinal barrier function in a mouse model of hyperuricemia. Mol Med Rep. 2019;20(4):3292-3300.
8. Yang Z, Ma H, Yin D, Sun C. Impact of Sacubitril/Valsartan on cardiac structure and blood levels of miRNA-328 and NT-proBNP in patients with CHD and chronic heart failure. Altern Ther Health Med. ;():-.
9. Jering KS, Claggett BL, Pfeffer MA. Prognostic importance of NT-proBNP (N-terminal pro-B-type natriuretic peptide) following high-risk myocardial infarction in the PARADISE-MI trial. Circ Heart Fail. 2023;16(5):e010259-.
10. Fan Y, Liang L, Tang X. Changes in the gut microbiota structure and function in rats with doxorubicin-induced heart failure. Front Cell Infect Microbiol. 2023;13():1135428-.
11. Emdin CA, Khera AV, Kathiresan S. Mendelian randomization. JAMA. 2017;318(19):1925-1926.
12. Visscher PM, Wray NR, Zhang Q. 10 years of GWAS discovery: biology, function, and translation. Am J Hum Genet. 2017;101(1):5-22.
13. Dan YL, Wang P, Cheng Z. Circulating adiponectin levels and systemic lupus erythematosus: a two-sample Mendelian randomization study. Rheumatol (Oxf Engl). 2021;60(2):940-946.
14. Long Y, Tang L, Zhou Y, Zhao S, Zhu H. Causal relationship between gut microbiota and cancers: a two-sample Mendelian randomisation study. BMC Med. 2023;21(1):66-.
15. Sanna S, van Zuydam NR, Mahajan A. Causal relationships among the gut microbiome, short-chain fatty acids and metabolic diseases. Nat Genet. 2019;51(4):600-605.
16. Kurilshikov A, Medina-Gomez C, Bacigalupe R. Large-scale association analyses identify host factors influencing human gut microbiome composition. Nat Genet. 2021;53(2):156-165.
17. Sun BB, Maranville JC, Peters JE. Genomic atlas of the human plasma proteome. Nature. 2018;558(7708):73-79.
18. Shen H, He Q, Shao X. Predictive value of NT-proBNP and hs-TnT for outcomes after pediatric congenital cardiac surgery. Int J Surg. 2024;110(6):3365-3372.
19. Zeng Q, Li D, He Y. Discrepant gut microbiota markers for the classification of obesity-related metabolic abnormalities. Sci Rep. 2019;9(1):13424-.
20. de Lemos JA, McGuire DK, Drazner MH. B-type natriuretic peptide in cardiovascular disease. Lancet (London, England). 2003;362(9380):316-322.
21. Vemuri R, Ruggiero A, Whitfield JM. Hypertension promotes microbial translocation and dysbiotic shifts in the fecal microbiome of nonhuman primates. Am J Physiol Heart Circ Physiol. 2022;322(3):H474-H485.
22. Wang L, Zhou W, Guo M. The gut microbiota is associated with clinical response to statin treatment in patients with coronary artery disease. Atherosclerosis. 2021;325():16-23.
23. Yang T, Qu H, Song X. Luhong granules prevent ventricular remodelling after myocardial infarction by reducing the metabolites TMAO and LPS of the intestinal flora. Evid Based Complement Alternat Med. 2019;2019():8937427-.
24. Zuo K, Li J, Li K. Disordered gut microbiota and alterations in metabolic patterns are associated with atrial fibrillation. GigaScience. 2019;8(6):giz058-.
25. Tian Y, Yao G, Skudder-Hill L. Gut microbiota’s causative relationship with peripheral artery disease: a Mendelian randomization study. Front Microbiol. 2024;15():1340262-.
26. Mao M, Zhai C, Qian G. Gut microbiome relationship with arrhythmias and conduction blocks: a two-sample Mendelian randomization study. J Electrocardiol. 2023;80():155-161.
27. Jiao J, Zhang Y, Han P, Zhai S. A preliminary study on the value of intestinal flora in predicting major adverse cardiovascular and cerebrovascular events in patients with refractory hypertension. Comp Math Methods Med. 2022;2022():7723105-.
28. Zhu Q, Gao R, Zhang Y. Dysbiosis signatures of gut microbiota in coronary artery disease. Physiol Genomics. 2018;50(10):893-903.
29. Zhu S, Gao B, Umetsu Y. Adaptively evolved human oral actinomyces-sourced defensins show therapeutic potential. EMBO Mol Med. 2022;14(2):e14499-.
30. Liu J, An N, Ma C. Correlation analysis of intestinal flora with hypertension. Exp Ther Med. 2018;16(3):2325-2330.
31. Han Y, Gong Z, Sun G. Dysbiosis of gut microbiota in patients with acute myocardial infarction. Front Microbiol. 2021;12():680101-.
32. Otto CM, Rahimi K. Heartbeat: the gut microbiota and heart failure. Heart (Br Card Soc). 2016;102(11):811-812.
33. Modrego J, Ortega-Hernández A, Goirigolzarri J. Gut microbiota and derived short-chain fatty acids are linked to evolution of heart failure patients. Int J Mol Sci. 2023;24(18):13892-.
34. van der Wal HH, Comin-Colet J, Klip IT. Vitamin B12 and folate deficiency in chronic heart failure. Heart (Br Card Soc). 2015;101(4):302-310.
35. Hu W, Yuan L, Wang X. Predictive value of arterial blood lactic acid concentration on the risk of in-hospital all-cause death in patients with acute heart failure. Int J Clin Pract. 2022;2022():7644535-.
36. Mayerhofer CCK, Ueland T, Broch K. Increased secondary/primary bile acid ratio in chronic heart failure. J Card Fail. 2017;23(9):666-671.
37. Jie Z, Xia H, Zhong SL. The gut microbiome in atherosclerotic cardiovascular disease. Nat Commun. 2017;8(1):845-.
38. Bartolomaeus H, Balogh A, Yakoub M. Short-chain fatty acid propionate protects from hypertensive cardiovascular damage. Circulation. 2019;139(11):1407-1421.
39. Haeusler RA, Astiarraga B, Camastra S, Accili D, Ferrannini E. Human insulin resistance is associated with increased plasma levels of 12α-hydroxylated bile acids. Diabetes. 2013;62(12):4184-4191.
40. Rahman MM, Islam F, -Or-Rashid MH. The gut microbiota (microbiome) in cardiovascular disease and its therapeutic regulation. Front Cell Infect Microbiol. 2022;12():903570-.