Skip to main content
Journal of the American Heart Association: Cardiovascular and Cerebrovascular Disease logoLink to Journal of the American Heart Association: Cardiovascular and Cerebrovascular Disease
. 2026 Sep 9;15(19):e051046. doi: 10.1161/JAHA.126.051046

Association of F11 Gene Variants With Cardiovascular Events and Cardiac Function

Qianxi Ye 1,#, Shengtao Hu 2,#, Rui Yang 2,3,#, Yilin Lai 1, Zhang Chen 1, Jiajun Li 1, Haige Zhao 1, Jiayuan Wu 4, Liang Ma 1,✉, Zhijun Dai 2,✉
PMCID: PMC13745127  PMID: 42757890

Abstract

Background

FXI (Factor XI) inhibitors have emerged as a promising antithrombotic strategy. However, a recent study linked decreased FXI levels with increased risk of heart failure (HF) and atrial fibrillation (AF), particularly in older adults. This study aims to investigate the associations of genetic FXI inhibition with cardiovascular events and cardiac function.

Methods

This prospective study included 408 642 UK Biobank participants of European ancestry. Two F11‐based genetic scores were constructed, weighted separately by their associations with activated partial thromboplastin time or circulating FXI levels. Adjusted Cox or linear regression models were built to predict incident HF, AF, and cardiac magnetic resonance measures. Interactions with age and age‐related cardiometabolic traits were tested.

Results

Over 14.6 years median follow‐up, 14 451 participants developed incident HF, and 29 243 developed incident AF. The FXI activated partial thromboplastin time score was not associated with HF, AF, or cardiac magnetic resonance measures overall. In exploratory interaction analyses, baseline hemoglobin A1c (HbA1c) modified its association with HF (P<0.001) and AF (P=0.027), with higher HF risk observed among participants with HbA1c≥7% (hazard ratio per 10% genetically predicted increase in activated partial thromboplastin time, 1.65 [95% CI, 1.13–2.40]) but not in those with HbA1c<7%. No association was observed with AF across HbA1c strata. Similar patterns were observed for the FXI level score.

Conclusions

Genetically proxied FXI inhibition was not associated with HF, AF, or cardiac magnetic resonance–derived cardiac measures overall, but its associations with HF and AF may be modified by HbA1c. Future studies are needed to verify these exploratory findings.

Keywords: atrial fibrillation, biological specimen banks, factor XI, fibrinolytic agents, follow‐up studies, glycated hemoglobin, glycemic control

Subject Categories: Heart Failure, Atrial Fibrillation, Anticoagulants


Nonstandard Abbreviations and Acronyms

aPTT

activated partial thromboplastin time

FXI

factor XI

GLS

global longitudinal strain

IS

ischemic stroke

Clinical Perspective.

What Is New?

  • In this large UK Biobank genetic study, genetically proxied FXI (factor XI) inhibition was not associated with incident heart failure, incident atrial fibrillation, or cardiac magnetic resonance–derived measures of cardiac structure and function overall.

  • Exploratory analyses suggested that baseline hemoglobin A1c may modify the associations of genetically proxied FXI inhibition with heart failure and atrial fibrillation, with a higher heart failure risk observed only among participants with hemoglobin A1c ≥7%.

What Are the Clinical Implications?

  • As FXI‐targeted therapies emerge as a promising class of novel anticoagulants, these findings support the need to evaluate the cardiovascular safety of FXI inhibitors in future studies, with consideration of glycemic control.

Observational epidemiologic studies have reported that high plasma FXI (factor XI) levels are an independent risk factor for venous thromboembolism (VTE) and ischemic stroke (IS) but individuals with congenital FXI deficiency rarely experience spontaneous bleeding. 1 , 2 , 3 , 4 , 5 Given these associations, therapeutic strategies targeting FXI/FXIa to mitigate bleeding while maintaining antithrombotic efficacy have attracted considerable interest. Currently, multiple FXI inhibitors are being evaluated in phase 3 clinical trials, involving >75 000 participants across diverse clinical settings, including long‐term antithrombotic therapy in patients with atrial fibrillation (AF), stroke, or cancer. 6 , 7 However, emerging evidence has raised concerns that FXI inhibition could pose cardiovascular risks. In a large community‐based study, lower circulating FXI levels were associated with a higher incidence of heart failure (HF) and a higher prevalence of AF in the older stratum (≥75 years) but not in the younger stratum. 8

Genetic variants, randomly allocated at conception, can serve as unconfounded proxies for protein perturbation to assess causal effects with a drug target Mendelian randomization approach. A previous Mendelian randomization study found no association between lifelong lower FXI levels and HF, its subtypes, AF, or cardiac magnetic resonance (CMR) measures of cardiac function in the general population. 9 These conflicting findings may be due to nonlinear risk relationships and effect heterogeneity across subgroups, which have not been fully explored. Previous studies have speculated that the age‐dependent association between FXI and HF may partly result from the higher prevalence of cardiometabolic traits, such as hypertension, with aging. 8 , 10 Recognizing potential age‐related risk profiles can inform therapeutic strategies.

Therefore, this study aimed to leverage the drug‐target Mendelian randomization paradigm to evaluate the association of F11 variants with incident HF, incident AF, and CMR‐derived cardiac structure and function using individual‐level data. Additionally, we assessed potential interactions between F11 variants, age, and age‐related cardiometabolic traits to identify subgroups where genetic FXI inhibition may increase cardiovascular risk.

METHODS

All data used in this analysis are available online at https://www.ukbiobank.ac.uk.

Study Population

The UKB (UK Biobank) is a large prospective cohort study that enrolled ∼500 000 participants aged 37 to 73 years between 2006 and 2010 across 22 assessment centers in England, Scotland, and Wales. The design and data collection procedures of UKB have been described previously. 11 Analyses were restricted to unrelated White British individuals with a kinship coefficient <0.0884. Participants with sex discordance, heterozygosity or genotype missingness outliers, or prevalent HF, AF, or structural heart disease were excluded as appropriate for each analysis. The detailed selection of participants for the study is shown in Figure 1. All participants provided written informed consent, and ethical approval was obtained from the North West Multi‐Center Research Ethics Committee. This research was conducted under the Application Number: 97101.

Figure 1. Flow diagram.

Figure 1

FXI Genetic Scores

We constructed 2 F11‐based genetic scores representing related but distinct proxies of FXI inhibition: an aPTT‐based score, considered the primary genetic instrument because aPTT is a clinically monitored pharmacodynamic marker of FXI inhibition and permits clinically interpretable scaling in its original unit, and a circulating FXI level–based score, used as a complementary protein‐level instrument. Genetic associations were derived from 9240 participants in the ARIC (Atherosclerosis Risk in Communities) study for aPTT (in seconds), 12 and from >35 000 participants in a deCODE genetics proteomic study for FXI levels (in SD units). 13

For each trait, we selected unambiguous, independent (linkage disequilibrium r 2<0.1) common variants (minor allele frequency>1%) reaching genome‐wide significance (P<5×10−8) within ±100 kb of the F11 locus. Genetic score construction details and complete variant lists are shown in Data S1 and Tables S1 and S2.

Clinical and Cardiac Imaging Outcomes

Incident HF and AF were the primary outcomes, with IS and VTE included as positive control outcomes. Outcomes were ascertained through linked electronic health records using International Classification of Diseases, Ninth Revision (ICD‐9) and Tenth Revision (ICD‐10), self‐reported noncancer illness codes, and hospital procedure codes. Incident events were identified from baseline through October 6, 2023, with follow‐up censored at the date of first event, death, loss to follow‐up, or end of follow‐up, whichever came first. Detailed outcome definitions are provided in Table S3.

CMR imaging was obtained from a subset of UKB participants at imaging visit beginning in 2014 using standardized protocols, as described previously. 14 Images were acquired on 1.5‐Tesla scanners (MAGNETOM Aera, Siemens Healthcare) and analyzed centrally using a standardized image analysis pipeline. We evaluated 6 cardiac structure and function measures, including indices of left atrial size and function (maximum and minimum left atrial volume, left atrial ejection fraction [EF]) and left ventricular structure and function (left ventricular mass, global longitudinal strain, and EF). 15 Body surface area–indexed measures were derived using the Du Bois formula.

Statistical Analysis

Cox proportional hazards models were used to assess associations of each FXI genetic score with incident clinical outcomes, and linear regression models were used for traditional cardiovascular risk factors and CMR traits. Details of the cardiovascular risk factors are presented in Data S1. All models were adjusted for age, sex, genotyping array, and the first 10 genetic principal components. The proportional hazards assumption was tested using Schoenfeld residuals. Potential nonlinearity in the associations between FXI genetic scores and outcomes was evaluated by comparing models with linear terms to those incorporating restricted cubic splines with 4 knots (5%, 35%, 65%, 95%), using likelihood ratio tests. The number of knots was determined based on Akaike's information criterion to identify the best fit while minimizing the risk of overfitting. Stratified analyses were prespecified by sex. Interaction terms were tested using likelihood ratio tests to compare models with and without interaction term.

Building on previous studies suggesting stronger associations between FXI and HF in older individuals, we prespecified age and 5 age‐related cardiometabolic traits, including hemoglobin A1c (HbA1c), systolic blood pressure, low‐density lipoprotein cholesterol, hs‐CRP (high‐sensitivity C‐reactive protein), and estimated glomerular filtration rate, as potential effect modifiers. 16 , 17 , 18 Nonlinear interaction analyses were conducted using restricted cubic splines with 4 knots, implemented via the InteractionRCS R package. The interaction splines were displayed as smooth curves, with the ends truncated for improved visual clarity. In the presence of a significant interaction, stratified analyses were then conducted using clinically relevant cut points.

Participants with missing data were excluded from the corresponding analyses; patterns of missingness are summarized in Table S4. P values were 2 sided, and a Bonferroni‐corrected significance threshold was applied for the 2 primary clinical outcomes (HF and AF; α=0.05/2) and for the 6 CMR measures (α=0.05/6). Interaction and subgroup analyses were considered hypothesis‐generating. All analyses were performed using R, version 4.3.1.

RESULTS

Baseline Characteristics

The incident HF and AF cohorts included 403 463 and 401 472 participants, respectively, with mean±SD ages of 56.9 (8.0) and 56.8 (8.0) years, and 45.8% and 45.5% men, respectively. Baseline characteristics were similar across FXI genetic score groups (Table and Table S5). During a median follow‐up of 14.6 (interquartile range, 13.7–15.3) years, 14 451 participants (3.6%) developed incident HF and 29 243 (7.3%) developed incident AF.

Table 1.

Participant Characteristics, FXI Genetic Scores, and Incident Heart Failure Analysis

Variable All FXI aPTT score FXI level score
Lower Higher P value Lower Higher P value
Participants, No. 403 463 202 481 200 982 201 733 201 730
Age, y, mean±SD 56.9±8.0 56.9±8.0 56.9±8.0 0.291 56.9±8.0 56.9±8.0 0.666
Sex, male, No. (%) 184 778 (45.8) 92 989 (45.9) 91 789 (45.7) 0.105 92 050 (45.6) 92 728 (46.0) 0.032
Current smoker, No. (%) 40 676 (10.1) 20 495 (10.2) 20, 181 (10.1) 0.393 20 424 (10.2) 20 252 (10.1) 0.369
Current drinker, No. (%) 376 915 (93.5) 189 130 (93.5) 187 785 (93.5) 0.736 188 527 (93.5) 188 388 (93.5) 0.353
Body mass index, kg/m2, mean±SD 27.4±4.7 27.4±4.7 27.4±4.7 0.152 27.4±4.7 27.4±4.7 0.553
Hypertension, No. (%) 107 192 (26.6) 53 676 (26.5) 53 516 (26.6) 0.398 53 755 (26.6) 53 437 (26.5) 0.260
Diabetes, No. (%) 22 390 (5.8) 11 287 (5.8) 11, 103 (5.8) 0.455 11, 152 (5.8) 11 238 (5.8) 0.549
Low‐density lipoprotein cholesterol, mg/dL, mean±SD 138.2±33.6 138.3±33.6 138.1±33.6 0.070 138.2±33.6 138.2±33.6 0.314
High‐sensitivity C‐reactive protein, mg/L, median (interquartile range) 1.3 (0.7–2.7) 1.3 (0.7–2.7) 1.3 (0.7–2.7) 0.277 1.3 (0.7–2.8) 1.3 (0.7–2.7) 0.146
Estimated glomerular filtration rate, mL/1.73 m2, mean±SD* 90.4±12.6 90.4±12.7 90.4±12.6 0.925 90.4±12.6 90.4±12.6 0.268

P values for continuous variables were calculated using t test and the Mann–Whitney U test, whereas those for categorical variables were determined using the chi‐square test. International System of Units conversion factor: To convert low‐density lipoprotein cholesterol to mmol/L, multiply by 0.0259. aPTT indicates activated partial thromboplastin time; and FXI, factor XI.

*

Calculated from serum creatinine using the Chronic Kidney Disease Epidemiology Collaboration equation.

FXI Genetic Scores and Validation

We identified 2 and 24 independent variants at the F11 locus for the FXI aPTT score and the FXI level score, explaining 16.6% and 17.6% of the variance in aPTT and FXI levels, respectively (F statistics 917 and 315). The FXI aPTT score was associated with 0.012% lower HbA1c (95% CI, −0.024% to −0.001%; P = 0.03) per 10% genetically predicted prolongation in aPTT, with no such association for the FXI level score. (Figure S1). Neither FXI genetic score was associated with lipids, blood pressure, body mass index, glucose, hs‐CRP, estimated glomerular filtration rate, smoking, or alcohol use. As expected, both scores were associated with lower risks of VTE and IS. (Figure 2).

Figure 2. Associations of FXI genetic scores with positive control outcomes, heart failure, and atrial fibrillation.

Figure 2

Cox proportional hazards models were adjusted for age, sex, genotyping array, and the first 10 genetic principal components. Associations are per 10% longer aPTT (A) or per 1‐SD lower FXI level (B), both genetically predicted. Squares represent hazard ratios; whiskers, 95% CIs. P values are not adjusted for multiple testing. Nonlinearity P values were derived by comparing linear models with restricted cubic spline models (4 knots) using likelihood ratio tests. aPTT indicates activated partial thromboplastin time; FXI, factor XI; and HR, hazard ratio.

FXI Genetic Scores and Cardiovascular Events

Neither FXI genetic score was associated with incident HF or AF (Figure 2). For HF, the HRs were 0.98 (95% CI, 0.88–1.08) per 10% genetically predicted aPTT prolongation and 0.99 (95% CI, 0.98–1.01) per 1‐SD lower FXI level. For AF, the corresponding HRs were 0.96 (95% CI, 0.89–1.03) and 0.99 (95% CI, 0.98–1.00). No evidence of nonlinearity or interaction with sex was observed (Table S6).

Impact of Age and Age‐Related Cardiometabolic Traits on the Association Between FXI Genetic Scores and Clinical Outcomes

Interaction analyses showed significant modification by HbA1c for both HF and AF, but not by age, systolic blood pressure, hs‐CRP, low‐density lipoprotein cholesterol, or estimated glomerular filtration rate (Figure 3 and Figures S2–S6). Baseline HbA1c was missing in 18 873 participants (4.68%) in the cohort with HF and 18 768 participants (4.67%) in the cohort with AF. In stratified analyses, FXI genetic scores were not associated with HF among participants with HbA1c <7% (HR per 10% genetically predicted increase in aPTT, 0.94 [95% CI, 0.84–1.04]; HR per 1‐SD lower FXI level, 0.99 [95% CI, 0.97–1.00]), whereas both scores were associated with higher HF risk among those with HbA1c ≥7% (HR, 1.65 [95% CI, 1.13–2.40]; HR, 1.08 [95% CI, 1.02–1.15], respectively). No association with AF was observed in either HbA1c stratum (Figure 4). Sensitivity analyses additionally adjusting for continuous HbA1c within strata yielded similar results (Figures S7). No comparable HbA1c interactions were observed for VTE or IS, and similar patterns were not observed for diabetes status (Figures S8 and S9).

Figure 3. Hazard ratio for heart failure and atrial fibrillation as a function of baseline HbA1c.

Figure 3

Risk for heart failure (A) or atrial fibrillation (B) is shown per 10% genetically predicted prolongation in aPTT. Risk for heart failure (C) or atrial fibrillation (D) is shown per 1‐SD genetically predicted lower FXI level. Shaded region represents 95% CI. P for interaction was obtained from likelihood ratio tests comparing models with and without the interaction term. aPTT indicates activated partial thromboplastin time; AF, atrial fibrillation; FXI, factor XI; HbA1c, hemoglobin A1c; HF, heart failure; and HR, hazard ratio.

Figure 4. Associations of FXI genetic scores with heart failure and atrial fibrillation stratified by baseline HbA1c.

Figure 4

Cox proportional hazards models were adjusted for age, sex, genotyping array, and the first 10 genetic principal components. Associations are per 10% longer aPTT (A) or per 1‐SD lower FXI level (B), both genetically predicted. Squares represent hazard ratios; whiskers, 95% CIs. P values are not adjusted for multiple testing. Nonlinearity P values were derived by comparing linear models with restricted cubic spline models (4 knots) using likelihood ratio tests. aPTT indicates activated partial thromboplastin time; FXI, factor XI; HbA1c, hemoglobin A1c; and HR, hazard ratio.

FXI Genetic Scores and Cardiac Structure and Function

Characteristics of participants for the analyses of CMR measures are displayed in Table S7. FXI genetic scores were not associated with any CMR measures of cardiac structure or function (Figure 5). Sex interactions with the FXI aPTT score were observed for maximum (P=0.043) and minimum left atrial volume (P=0.042), but stratified analyses did not show significant associations (Table S8). No interactions with age, systolic blood pressure, hs‐CRP, low‐density lipoprotein cholesterol, HbA1c, or estimated glomerular filtration rate were observed (Figure S10).

Figure 5. Associations of FXI genetic scores with cardiac magnetic resonance measures.

Figure 5

Linear regression models were adjusted for age, sex, genotyping array, and the first 10 genetic principal components. Associations are per 10% longer aPTT (A) or per 1‐SD lower FXI level (B), both genetically predicted. P values are not adjusted for multiple testing. Nonlinearity P values were derived by comparing linear models with restricted cubic spline models (4 knots) using likelihood ratio tests. aPTT indicates activated partial thromboplastin time; CMR, cardiac magnetic resonance; FXI, factor XI; LA, left atrial; and LV, left ventricular.

DISCUSSION

In this large genetic study, genetic FXI inhibition, indexed by prolonged aPTT or lower FXI levels, was not associated with incident HF, incident AF, or CMR measures of cardiac structure and function in the general population. As expected, FXI genetic scores were associated with a lower risk of VTE and IS. Exploratory interaction analyses suggested that the association between FXI genetic scores and HF and AF risk was modified by baseline HbA1c. In stratified analyses, FXI genetic scores were associated with higher HF risk among individuals with HbA1c ≥7%, whereas no association was observed among those with HbA1c <7% or for AF.

These overall null findings align with recent Mendelian randomization evidence showing no association of genetically lower FXI with HF, AF, or CMR traits. 9 Lower plasma FXI levels were also not associated with greater HF or AF risk in a prospective study involving 11 766 Black and White participants (median age, 60 years), similar in age to our study population. 8 Although congenital FXI deficiency is well documented in relation to thromboembolic events, evidence linking it to HF or AF remains limited, including in high‐prevalence populations such as Ashkenazi Jews. 19 , 20 Collectively, available evidence does not support a broad adverse association between FXI inhibition and cardiovascular outcomes in the general population.

Our exploratory findings suggest possible heterogeneity in FXI–cardiac outcome associations across metabolic contexts. The absence of similar interactions with diabetes status may reflect the greater statistical power of HbA1c as a continuous measure and the potential relevance of glycemic control beyond diabetes diagnosis alone. No comparable interactions were observed for VTE or IS. These findings may provide a possible context for interpreting prior studies reporting adverse associations between lower FXI levels and HF. In humans, adverse associations between lower FXI levels and HF have been observed only in older populations, 8 , 21 among whom suboptimal glycemic control is common and a substantial proportion of individuals with diabetes have HbA1c ≥7.0%. 22 , 23 , 24 Previous research on HF with preserved EF in mice demonstrated that FXI activates the bone morphogenetic protein–SMAD1/5 pathway and inhibits inflammatory and fibrotic genes, thereby preserving diastolic function. 25 Notably, the “2‐hit” HF with preserved EF model, induced by high‐fat diet and nitric oxide synthase inhibition, also causes metabolic disorders, including elevated glycated hemoglobin. 26 , 27 However, because the HbA1c‐related results arose from exploratory analyses, and because our study did not directly assess the biological pathways linking FXI, glycemic status, myocardial remodeling, and HF subtypes, these interpretations remain speculative and require validation in future clinical and mechanistic studies.

Previous observational evidence linked lower FXI levels with prevalent AF, particularly in older populations but not with incident AF. 8 In our study, genetically proxied FXI inhibition was not associated with incident AF overall. Although exploratory analyses suggested effect modification by HbA1c, no significant association was observed in either HbA1c stratum, possibly reflecting a weaker continuous AF pattern not fully captured by a single clinical cutoff. These findings therefore warrant cautious interpretation. Given the ongoing evaluation of FXI inhibitors for stroke prevention in patients with AF, future studies are needed to clarify the relationship between FXI‐related pathways and AF. 28

In addition, CMR analyses did not provide evidence that genetically proxied FXI inhibition was associated with cardiac structural or functional alterations, and no significant interactions with age or age‐related cardiometabolic traits were observed. Nevertheless, the smaller imaging sample size and cross‐sectional assessment may have limited detection of subtle or longitudinal changes in cardiac remodeling.

This study has several strengths. Plasma FXI levels are relatively stable throughout life, 29 and the use of cis F11 genetic scores as proxies for lifelong FXI perturbation reduces susceptibility to reverse causation and confounding compared with conventional observational analyses. In addition, the aPTT‐based score may better reflect the clinically monitored anticoagulant activity of FXI inhibition, reducing potential misclassification among individuals with low FXI antigen but preserved coagulation function. 30 Several limitations should be noted. First, interaction analyses across multiple cardiometabolic traits, outcomes, and 2 FXI genetic scores were considered hypothesis‐generating and were not adjusted for multiple testing; therefore, the HbA1c‐related findings may reflect chance and require further validation. Second, outcome ascertainment relied on linked health records, self‐reported diagnoses, and procedural codes rather than systematic screening, which may have led to underdetection of asymptomatic or paroxysmal AF and incomplete capture of outpatient HF or HF subtypes. Greater health care contact among participants with higher HbA1c could also have contributed to differential outcome ascertainment across HbA1c strata. Third, the White British ancestry of the UKB cohort limits the generalizability of the findings to more diverse populations. Fourth, cardiometabolic traits were measured only at baseline, precluding assessment of long‐term glycemic control or temporal changes. Finally, genetic proxies represent lifelong, modest FXI perturbation and are not equivalent to pharmacologic FXI inhibition, whose effects are dose and time dependent and may differ from developmental or compensatory effects of lifelong F11 variation.

CONCLUSIONS

In conclusion, genetically proxied FXI inhibition was not associated with incident HF, incident AF, or CMR‐derived cardiac measures overall. Exploratory analyses suggested that HbA1c may modify its associations with HF and AF. These findings require further validation and support consideration of glycemic control in future studies evaluating the cardiovascular safety of FXI‐targeted therapies.

Sources of Funding

This study was supported by grant 82500490 from the National Natural Science Foundation of China (Zhang Chen); grant QN26H100010 from the Natural Science Foundation of Zhejiang Province (Zhang Chen); and grant 2025M782267 from the China Postdoctoral Science Foundation (Zhang Chen).

Disclosures

None.

Supporting information

Supplemental Methods

Tables S1–S8

Figures S1–S10

References 31–35

STROBE Statement—checklist

JAH3-15-e051046-s002.pdf (151.6KB, pdf)

Acknowledgments

We thank all the participants of the UK Biobank and all the people involved in building the large biobank. Zhijun Dai, Jiayuan Wu, and Rui Yang had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: Qianxi Ye, Shengtao Hu; acquisition, analysis, or interpretation of data: Shengtao Hu, Rui Yang, Jiayuan Wu; drafting of the article: Qianxi Ye; critical review of the article for important intellectual content: all authors; statistical analysis: Qianxi Ye, Yilin Lai, Jiajun Li, Rui Yang; obtained funding: Zhang Chen; administrative, technical, or material support: Jiayuan Wu, Liang Ma, Zhijun Dai; supervision: Jiayuan Wu, Liang Ma, Zhijun Dai.

This article was sent to Jacquelyn Y. Taylor, PhD, PNP‐BC, RN, FAHA, FAAN, Associate Editor, for review by expert referees, editorial decision, and final disposition.

For Sources of Funding and Disclosures, see page 9.

Contributor Information

Liang Ma, Email: ml1402@zju.edu.cn.

Zhijun Dai, Email: dzj0911@zju.edu.cn.

References

  • 1. Bolton‐Maggs PH. Factor XI deficiency and its management. Haemophilia. 2000;6:100–109. doi: 10.1046/j.1365-2516.2000.00053.x [DOI] [PubMed] [Google Scholar]
  • 2. Meijers JC, Tekelenburg WL, Bouma BN, Bertina RM, Rosendaal FR. High levels of coagulation factor XI as a risk factor for venous thrombosis. N Engl J Med. 2000;342:696–701. doi: 10.1056/nejm200003093421004 [DOI] [PubMed] [Google Scholar]
  • 3. Paszek E, Polak M, Undas A. Elevated factor XI is associated with increased risk of recurrent cerebral venous sinus thrombosis: a cohort study. J Thromb Thrombolysis. 2024;57:402–407. doi: 10.1007/s11239-023-02935-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Cushman M, O'Meara ES, Folsom AR, Heckbert SR. Coagulation factors IX through XIII and the risk of future venous thrombosis: the Longitudinal Investigation of Thromboembolism Etiology. Blood. 2009;114:2878–2883. doi: 10.1182/blood-2009-05-219915 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Siegerink B, Rosendaal FR, Algra A. Antigen levels of coagulation factor XII, coagulation factor XI and prekallikrein, and the risk of myocardial infarction and ischemic stroke in young women. J Thromb Haemost. 2014;12:606–613. doi: 10.1111/jth.12531 [DOI] [PubMed] [Google Scholar]
  • 6. Gailani D, Gruber A. Targeting factor XI and factor XIa to prevent thrombosis. Blood. 2024;143:1465–1475. doi: 10.1182/blood.2023020722 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Capodanno D, Alexander JH, Bahit MC, Eikelboom JW, Gibson CM, Goodman SG, Kunadian V, Lip GYH, Lopes RD, Mehran R, et al. Factor XI inhibitors for the prevention and treatment of venous and arterial thromboembolism. Nat Rev Cardiol. 2025;22:896–912. doi: 10.1038/s41569-025-01144-z [DOI] [PubMed] [Google Scholar]
  • 8. Ji Y, Zhang MJ, Wang W, Norby FL, Eaton AA, Inciardi RM, Alonso A, Sedaghat S, Ganz P, Van't Hof J, et al. Association of coagulation factor XI level with cardiovascular events and cardiac function in community‐dwelling adults: from ARIC and CHS. Circulation. 2025;151:356–367. doi: 10.1161/circulationaha.124.070278 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Daghlas I, Gill D. Human genetic evidence does not support a causal role for reduced FXI levels in heart failure pathogenesis‐brief report. Arterioscler Thromb Vasc Biol. 2025;45:1664–1669. doi: 10.1161/atvbaha.125.322971 [DOI] [PubMed] [Google Scholar]
  • 10. Feng S, Wang Z, Sun R. Letter by Feng et al regarding article, “Association of coagulation factor XI level with cardiovascular events and cardiac function in community‐dwelling adults: from ARIC and CHS”. Circulation. 2025;152:e239–e240. doi: 10.1161/circulationaha.124.073382 [DOI] [PubMed] [Google Scholar]
  • 11. Sudlow C, Gallacher J, Allen N, Beral V, Burton P, Danesh J, Downey P, Elliott P, Green J, Landray M, et al. UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 2015;12:e1001779. doi: 10.1371/journal.pmed.1001779 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Tang W, Schwienbacher C, Lopez LM, Ben‐Shlomo Y, Oudot‐Mellakh T, Johnson AD, Samani NJ, Basu S, Gögele M, Davies G, et al. Genetic associations for activated partial thromboplastin time and prothrombin time, their gene expression profiles, and risk of coronary artery disease. Am J Hum Genet. 2012;91:152–162. doi: 10.1016/j.ajhg.2012.05.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Ferkingstad E, Sulem P, Atlason BA, Sveinbjornsson G, Magnusson MI, Styrmisdottir EL, Gunnarsdottir K, Helgason A, Oddsson A, Halldorsson BV, et al. Large‐scale integration of the plasma proteome with genetics and disease. Nat Genet. 2021;53:1712–1721. doi: 10.1038/s41588-021-00978-w [DOI] [PubMed] [Google Scholar]
  • 14. Petersen SE, Matthews PM, Francis JM, Robson MD, Zemrak F, Boubertakh R, Young AA, Hudson S, Weale P, Garratt S, et al. UK Biobank's cardiovascular magnetic resonance protocol. J Cardiovasc Magn Reson. 2016;18:8. doi: 10.1186/s12968-016-0227-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Chamsi‐Pasha MA, Zhan Y, Debs D, Shah DJ. CMR in the evaluation of diastolic dysfunction and phenotyping of HFpEF: current role and future perspectives. JACC Cardiovasc Imaging. 2020;13:283–296. doi: 10.1016/j.jcmg.2019.02.031 [DOI] [PubMed] [Google Scholar]
  • 16. Murakata Y, Fujimaki T, Yamada Y. Age‐related changes in clinical parameters and their associations with common complex diseases. Biomed Rep. 2015;3:767–777. doi: 10.3892/br.2015.505 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Singh GM, Danaei G, Pelizzari PM, Lin JK, Cowan MJ, Stevens GA, Farzadfar F, Khang YH, Lu Y, Riley LM, et al. The age associations of blood pressure, cholesterol, and glucose: analysis of health examination surveys from international populations. Circulation. 2012;125:2204–2211. doi: 10.1161/circulationaha.111.058834 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Li T, Chen N, Liu Z, Shan Z, Dong G, Yang J, Qi M. Age‐related differences in the association between plasma high‐sensitivity C‐reactive protein and noncalcified or mixed coronary atherosclerotic plaques. Mediat Inflamm. 2020;2020:5938957. doi: 10.1155/2020/5938957 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Lova A, Pagán J, de la Morena G, Vázquez DJ, Cerezo‐Manchado JJ, Bravo‐Pérez C, Miñano A, Tomás A, Vicente V, Lozano ML, et al. Congenital factor XI deficiency and risk of heart failure in humans. J Thromb Haemost. 2023;21:2626–2629. doi: 10.1016/j.jtha.2023.06.013 [DOI] [PubMed] [Google Scholar]
  • 20. Asselta R, Paraboschi EM, Rimoldi V, Menegatti M, Peyvandi F, Salomon O, Duga S. Exploring the global landscape of genetic variation in coagulation factor XI deficiency. Blood. 2017;130:e1–e6. doi: 10.1182/blood-2017-04-780148 [DOI] [PubMed] [Google Scholar]
  • 21. Alehagen U, Dahlström U, Lindahl TL. Low plasma concentrations of coagulation factors II, VII and XI indicate increased risk among elderly with symptoms of heart failure. Blood Coagul Fibrinolysis. 2010;21:62–69. doi: 10.1097/MBC.0b013e328332aa2b [DOI] [PubMed] [Google Scholar]
  • 22. Wang L, Li X, Wang Z, Bancks MP, Carnethon MR, Greenland P, Feng YQ, Wang H, Zhong VW. Trends in prevalence of diabetes and control of risk factors in diabetes among US adults, 1999‐2018. JAMA. 2021;326:1–13. doi: 10.1001/jama.2021.9883 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Wang L, Peng W, Zhao Z, Zhang M, Shi Z, Song Z, Zhang X, Li C, Huang Z, Sun X, et al. Prevalence and treatment of diabetes in China, 2013‐2018. JAMA. 2021;326:2498–2506. doi: 10.1001/jama.2021.22208 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Mata‐Cases M, Mauricio D, Real J, Vlacho B, Romera‐Liebana L, Molist‐Brunet N, Cedenilla M, Franch‐Nadal J. Potential risk of overtreatment in patients with type 2 diabetes aged 75 years or older: data from a population database in Catalonia, Spain. J Clin Med. 2022;11:11. doi: 10.3390/jcm11175134 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Cao Y, Wang Y, Zhou Z, Pan C, Jiang L, Zhou Z, Meng Y, Charugundla S, Li T, Allayee H, et al. Liver‐heart cross‐talk mediated by coagulation factor XI protects against heart failure. Science. 2022;377:1399–1406. doi: 10.1126/science.abn0910 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Schiattarella GG, Altamirano F, Tong D, French KM, Villalobos E, Kim SY, Luo X, Jiang N, May HI, Wang ZV, et al. Nitrosative stress drives heart failure with preserved ejection fraction. Nature. 2019;568:351–356. doi: 10.1038/s41586-019-1100-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Cordero‐Herrera I, Guimarães DD, Moretti C, Zhuge Z, Han H, McCann Haworth S, Uribe Gonzalez AE, Andersson DC, Weitzberg E, Lundberg JO, et al. Head‐to‐head comparison of inorganic nitrate and metformin in a mouse model of cardiometabolic disease. Nitric Oxide. 2020;97:48–56. doi: 10.1016/j.niox.2020.01.013 [DOI] [PubMed] [Google Scholar]
  • 28. Jain SS, Mahaffey KW, Pieper KS, Shimizu W, Potpara T, Ruff CT, Kamel H, Lewis BS, Cornel JH, Kowey PR, et al. Milvexian vs apixaban for stroke prevention in atrial fibrillation: the LIBREXIA atrial fibrillation trial rationale and design. Am Heart J. 2024;277:145–158. doi: 10.1016/j.ahj.2024.08.011 [DOI] [PubMed] [Google Scholar]
  • 29. Barg AA, Livnat T, Kenet G. Factor XI deficiency: phenotypic age‐related considerations and clinical approach towards bleeding risk assessment. Blood. 2024;143:1455–1464. doi: 10.1182/blood.2023020721 [DOI] [PubMed] [Google Scholar]
  • 30. Wheeler AP, Gailani D. Why factor XI deficiency is a clinical concern. Expert Rev Hematol. 2016;9:629–637. doi: 10.1080/17474086.2016.1191944 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Büller HR, Bethune C, Bhanot S, Gailani D, Monia BP, Raskob GE, Segers A, Verhamme P, Weitz JI. Factor XI antisense oligonucleotide for prevention of venous thrombosis. N Engl J Med. 2015;372:232–240. doi: 10.1056/NEJMoa1405760 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Weitz JI, Tankó LB, Floege J, Fox KAA, Bhatt DL, Thadhani R, Hung J, Pap ÁF, Kubitza D, Winkelmayer WC. Anticoagulation with osocimab in patients with kidney failure undergoing hemodialysis: a randomized phase 2 trial. Nat Med. 2024;30:435–442. doi: 10.1038/s41591-023-02794-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Palmer TM, Lawlor DA, Harbord RM, Sheehan NA, Tobias JH, Timpson NJ, Davey Smith G, Sterne JA. Using multiple genetic variants as instrumental variables for modifiable risk factors. Stat Methods Med Res. 2012;21:223–242. doi: 10.1177/0962280210394459 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF III, Feldman HI, Kusek JW, Eggers P, Van Lente F, Greene T, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150:604–612. doi: 10.7326/0003-4819-150-9-200905050-00006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Wang D, Morton JI, Salim A, Shaw JE, Magliano DJ. Association between diabetes and anemia: evidence from NHANES and the UK biobank. Diabetes Care. 2025;48:816–826. doi: 10.2337/dc24-2535 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Methods

Tables S1–S8

Figures S1–S10

References 31–35

STROBE Statement—checklist

JAH3-15-e051046-s002.pdf (151.6KB, pdf)

Articles from Journal of the American Heart Association: Cardiovascular and Cerebrovascular Disease are provided here courtesy of Wiley

RESOURCES