Abstract
Introduction:
Association between arterial vascular dysfunction and risk of venous thromboembolism (VTE) is uncertain. We determined the associations between comprehensive measures of arterial vascular function and risk of incident VTE in a community-based cohort study with robust longitudinal follow-up.
Materials and Methods:
In the Framingham Heart Study Original, Offspring, Third Generation, and Omni cohorts, we measured carotid-femoral pulse wave velocity and central pulse pressure (n=8261, age 51.5±15.5 years, 54% women), flow-mediated dilation and hyperemic velocity (n=6540, age 47.9±14.1 years, 54% women), and peripheral arterial tonometry ratio (n=4998, age 54.3±16.0 years, 52% women). Deep venous thrombosis and pulmonary embolism were diagnosed with imaging studies and adjudicated by three Framingham Heart Study physicians.
Results and Conclusions:
The rate of incident VTE was 1.6–2.1 per 1000 person-years during mean follow-up of 8.5–11.2 years. In age- and sex-adjusted Cox proportional hazards regression models, carotid-femoral pulse wave velocity was associated with increased risk of VTE (HR 1.32, 95% CI 1.05–1.66, p=0.02), however the association was no longer statistically significant after multivariable adjustment (HR 1.24, 95% CI 0.96–1.61, p=0.10). None of the other vascular variables were associated with the risk of VTE in any of the models. In our comprehensive examination of arterial vascular function and risk of VTE, we did not observe any association between select arterial function measures and risk of VTE after multivariable adjustment.
Keywords: venous thromboembolism, arterial stiffness, endothelial function
Introduction
Venous thromboembolism (VTE) is an important source of public health burden, affecting 300,000 – 600,000 individuals,[1] and accounting for more than 100,000 deaths annually in the US.[2,3] Classic risk factors for VTE include prolonged immobilization, major surgery, significant trauma, active cancer, inherited thrombophilia, and pregnancy, which together account for about two-thirds of all VTE cases.[4] About 30% of incident VTE cases are not associated with a clear preceding triggering event and remain unexplained after diagnostic workup.[4,5]
To date, potential relations between arterial vascular dysfunction and the risk of VTE have been largely unexplored in prospective longitudinal studies.[6] Although arterial and venous thromboembolism are caused by distinct pathologic mechanisms, there are several reasons to hypothesize that arterial stiffness and arterial endothelial dysfunction may be associated with increased risk of VTE. In prior cross-sectional studies, patients with VTE had increased arterial stiffness compared to those without VTE.[7,8] Atherosclerosis and VTE share common cardiovascular risk factors,[9–11] and there is a higher prevalence of subclinical atherosclerosis in patients with unprovoked VTE compared to those without VTE.[5] In addition, coronary microvascular endothelial dysfunction is associated with higher risk of VTE, which suggests that endothelial dysfunction may be an underlying mechanism for both arterial thrombosis and VTE.[6] In the current study, we aimed to examine associations between measures of arterial vascular function and risk of incident VTE in a prospective community-based cohort study.
Materials and Methods
Study Samples
Descriptions of the study design and the cohorts of the Framingham Heart Study (FHS) have been provided in detail previously.[12] In brief, the Original cohort was established in 1948. The Offspring cohort was established in 1971 and includes the children (and their spouses) of the Original Cohort. The Third Generation cohort was established in 2002 and includes adults aged ≥ 20 years with at least one parent in the Offspring cohort. The Omni-1 cohort was started in 1995 to reflect increased racial and ethnic diversity of the town of Framingham since the inception of the Original cohort. The Omni-2 cohort was begun in 2003, and some of its participants include family members of the Omni-1 cohort. Each cohort undergoes in-person examinations (i.e., exam cycles) every 2–7 years.
Carotid-femoral pulse wave velocity (CFPWV) and central pulse pressure (CPP) were measured in the Original cohort at exam 26 (1999–2001), in the Offspring cohort exams 7–9 (1998–2014), in the Omni-1 cohort at exams 2–4 (1999–2014), and the Third Generation and Omni-2 cohorts at exams 1–2 (2002–2011). Flow-mediated dilation (FMD) and hyperemic velocity were measured in the Original cohort at exam 28 (2004–2005), the Offspring cohort at exam 7 (1998–2001(, the Third Generation and Omni-2 cohorts at exam 1 (2002–2005). Peripheral arterial tonometry (PAT) ratio was measured in the Original cohort at exam 28 (2004–005), in the Offspring cohort at exam 8 (2005–2008), the Omni-1 cohort at exam 3 (2007–2008) and in the Third Generation and Omni-2 cohorts at exam 1 (2002–2005). If a participant attended more than one exam where a given vascular function variable was measured, their first exam was selected. Participants with missing data on the hemodynamic variables, missing covariates, lack of follow-up for VTE, or prevalent VTE at the time of baseline exam were excluded from the study. Details of the study sample selection are shown in Figure 1. The Institutional Review Boards of Boston University Medical Center approved the study, and all participants provided written informed consent.
Figure 1.
Flow diagram of study sample selection
Abbreviations: CFPWV, carotid-femoral pulse wave velocity; CPP, central pulse pressure; FMD, flow-mediated dilation; PAT, peripheral arterial tonometry; VTE, venous thromboembolism
Acquisition of CFPWV and CPP
Before all vascular tests, participants were instructed to not eat or drink anything other than water or decaffeinated coffee or tea after 8 PM the night before. Arterial tonometry was obtained from the brachial, radial, femoral, and carotid arteries in the supine position along with a simultaneously acquired ECG using a custom transducer and data acquisition system (NIHem Hemodynamic Workstation, Cardiovascular Engineering, Inc., Norwood, MA). The carotid-femoral transit distance was defined as the difference in body surface distances from the suprasternal notch to the carotid and femoral arteries. CFPWV was calculated by dividing this distance by the carotid-femoral pulse wave transit time.
To obtain CPP, we first calibrated the peak and trough of the signal-averaged brachial waveform using the brachial cuff systolic and diastolic blood pressures measured at the time of tonometry acquisition. Oscillometric blood pressure was used for Offspring/Omni-1 exam 7/2 and auscultatory blood pressure was used for all other exams. We calibrated the carotid pressure waveforms using the diastolic and integrated mean brachial pressures and defined CPP as the difference between the peak and trough of the calibrated carotid pressure waveform.
Acquisition of FMD and hyperemic flow velocity in brachial artery
Methodology for measuring brachial artery diameters and flow has been described previously.[13–15] Briefly, high-resolution ultrasonography was used (Toshiba SSH-140A, 7.5-MHz linear array transducer in Offspring and Omni-1 and Philips Sonos 5500, 11- to 3-L linear array transducer in Third Generation and Omni-2) to image brachial artery at rest and at 1 minute after induction of reactive hyperemia by 5 minutes of forearm cuff-occlusion. We measured brachial artery diameter using commercially available software (Brachial Analyzer; Medical Imaging Applications, Iowa City, IA; Version 3.2.3.sp2). We measured brachial artery flow using pulsed Doppler at rest and after cuff release for 15 seconds. We calculated FMD as the percent change in brachial diameter from rest to hyperemia (100*[hyperemic diameter at 60 seconds–resting diameter]/resting diameter). Doppler flow recordings were signal-averaged using the electrocardiographic R-wave as a fiducial point and were corrected for insonation angle. Integrated Doppler waveforms were used to calculate mean flow velocities at baseline and during hyperemia. Variability in measurements of tonometry [14] and FMD [13] from FHS were reported previously.
Acquisition of digital PAT
Methodology for acquisition of PAT has been described previously.[15,16] Briefly, we used a PAT device (Endo-PAT2000; Itamar Medical, Caesarea, Israel) to measure pulse amplitude at rest in the fingertip of the index finger of each hand. We then induced reactive hyperemia in one arm with 5 minutes of forearm cuff-occlusion. Pulse amplitude was recorded in both arms and data were analyzed by a computerized algorithm (Itamar Medical). To calculate the PAT ratio, the ratio of the post-deflation pulse amplitude to the baseline pulse amplitude in the 90- to 120-second post-deflation time period (i.e., Xh90–120/ Xh0: h denotes hyperemic finger, 0 denotes baseline, and X represents the pulse amplitude) was divided by the corresponding ratio from the contralateral hand (i.e., Xc90–120/ Xc0: with c denoting the control finger and 0 denoting baseline).
VTE Ascertainment
Diagnosis of VTE was made based on combination of symptoms suggestive of VTE, and imaging data.[17] Medical history was obtained during routine study visits (every 2 years for Original cohort, every 4–8 years for Offspring, Third Generation, Omni-1, and Omni-2 cohorts), and medical records were obtained for each reported outpatient visit and hospital admission related to symptoms suggestive of VTE. Pulmonary embolism was diagnosed by using either ventilation perfusion scan or computed tomography pulmonary angiogram. Deep venous thrombosis (DVT) was diagnosed by using one or more of the following diagnostic tests: a venogram, Doppler ultrasound, and 125I fibrinogen leg scan, impedance plethysmography or a combination of impedance plethysmography and leg radionuclide scans. We excluded thrombosis in the superficial veins and isolated DVT in the distal calf. The pulmonary embolism and DVT diagnoses were adjudicated by a panel of three Framingham Heart Study physicians reviewing external medical records.
We stratified VTE events as provoked or unprovoked based on previously used criteria. [17] We considered a VTE to be provoked if it was preceded by the following risk factor within 3 months prior to the VTE event: any surgery, any fracture/trauma requiring medical attention, hospitalization, documented bed rest ≥ 2 days, hormone replacement therapy or oral contraceptive use, being pregnant or up to 3 months postpartum at the time of VTE event, and travel >4 hours within 1 month prior.
Statistical Analysis
Summary statistics for all variables were calculated by using the mean (standard deviation), median (25th percentile, 75th percentile), or frequency counts (percentages), as appropriate. Due to skewness, both PAT ratio and triglycerides were natural logarithm (ln) transformed. Similarly, the distribution of CFPWV was transformed by taking the inverse and multiplying by −1000 in order to convert units to ms/m and maintain the original directionality of the variable (higher value represents a stiffer aorta).
Participants were followed until the date of VTE diagnosis, last exam attended, death, or December 31, 2014, whichever occurred first. In our primary analysis, Cox proportional hazards regression models were constructed to calculate hazard ratios (HR) and their 95% confidence intervals (CI) for the associations between CFPWV, CPP, FMD, hyperemic velocity, PAT ratio, and incident VTE. All models were adjusted for age and sex and stratified by cohort membership. A multivariable-adjusted model was constructed with further adjustment for mean arterial pressure (CFPWV and CPP only), systolic and diastolic blood pressures (FMD, hyperemic velocity, and PAT ratio only), heart rate, anti-hypertensive medication use, height, weight, diabetes, current smoking, history of myocardial infarction, history of heart failure, total/HDL cholesterol ratio, triglycerides, and lipid lowering medication use. Consistent with prior studies, we selected these variables for multivariable adjustment as they are associated with the vascular function measurements. [14–16] The proportional hazards assumption was verified using the Kolmogorov-Type Supremum Test for Proportional Hazards, from the ASSESS statement in SAS’s PHREG procedure.[18] All variables satisfied the proportional hazards assumption.
In a secondary analysis, we repeated our primary analysis excluding participants with a history of myocardial infarction or heart failure at baseline. Additionally, we assessed the presence of interaction between each of the vascular measures and the following variables: age, sex, hypertension, and diabetes. The interaction was tested by including a cross-product term in the multivariable-adjusted model and assessing its significance using a Chi-square test. All analyses were performed using SAS version 9.4 (Cary, NC). A two-sided p-value of <0.05 was considered statistically significant.
Results
The descriptive characteristics of the three study samples are provided in Table 1. The mean age of the participants were 47–54 years and about half were women. The rate of incident VTE was 1.6–2.1 per 1000 person-years, with a mean follow-up of 8.5–11.2 years. The mean age at the time of VTE diagnosis was 67.5–70.3 years. Table 2 reports numbers of events and HR for associations of arterial stiffness and endothelial dysfunction variables with incident VTE. CFPWV was significantly associated with increased risk for incident VTE after age and sex adjustment, but the association was no longer statistically significant after adjusting for additional clinical factors (Supplemental Figure 1). None of the other variables were associated with the risk of VTE in either the age and sex-adjusted or the multivariable-adjusted models. In the secondary analyses excluding prevalent cardiovascular diseases (Table 3) or prevalent myocardial infarction and congestive heart failure (Supplemental Table 1), the results were similar to those in our primary analysis. In our interaction analyses, we observed a statistically significant interaction between CPP and hypertension (p-value for interaction=0.04). Among participants with hypertension (N=2740, 101 VTE events), the HR for a 1 SD increment of CPP was 1.02 (95% CI: 0.83–1.27, p-value=0.83), while among participants without hypertension (N=5521, 84 VTE events), the HR was 0.77 (95% CI: 0.55–1.09, p-value=0.14). There were no other statistically significant interactions for any of the other variables.
Table 1.
Study sample characteristics
| Participants: |
Sample #1
Both CFPWV and CPP (N=8261) |
Sample #2
Both FMD and Hyperemic Velocity (N=6540) |
Sample #3
PAT (N=4998) |
|---|---|---|---|
| Age, years | 51.5±15.5 | 47.9±14.1 | 54.5±16.0 |
| Women | 4480 (54.2) | 3511 (53.7) | 2587 (51.8) |
| Race | |||
| White | 6953 (84.2) | 5708 (87.3) | 4352 (87.1) |
| Black | 295 (3.6) | 75 (1.2) | 201 (4.0) |
| Hispanic | 235 (2.8) | 152 (2.3) | 181 (3.6) |
| Asian | 174 (2.1) | 101 (1.5) | 153 (3.1) |
| Native Hawaiian/Pacific Islander | 27 (0.3) | 2 (0.03) | 3 (0.06) |
| American Indian/Alaska Native | 6 (0.1) | 4 (0.06) | 5 (0.1) |
| Multiple races | 15 (0.2) | 11 (0.17) | 14 (0.3) |
| Unknown | 556 (6.7) | 487 (7.5) | 89 (1.8) |
| Years of follow-up | 10.6±3.2 | 11.2±2.8 | 8.5±2.0 |
| FHS cohort | |||
| Height, cm | 168±10 | 169±10 | 169±10 |
| Weight, kg | 77.7±17.7 | 78.3±18.2 | 79.1±18.3 |
| Body mass index, kg/m2 | 27.3±5.2 | 27.2±5.4 | 27.7±5.5 |
| Diabetes mellitus | 597 (7.2) | 378 (5.8) | 459 (9.2) |
| Current smoking | 1048 (12.7) | 932 (14.3) | 577 (11.5) |
| Total cholesterol, mg/dL | 191±37 | 193±37 | 187±37 |
| HDL cholesterol, mg/dL | 55±17 | 54±17 | 55±17 |
| Triglycerides, mg/dL | 100 [70, 148] | 99 [69, 149] | 99 [70, 142] |
| Cholesterol treatment | 1352 (16.4) | 787 (12.0) | 1341 (26.8) |
| Heart rate, beats/min | 62±10 | 63±10 | 62±9 |
| SBP, mm Hg | 122±18 | 121±17 | 123±17 |
| DBP, mm Hg | 74±10 | 75±10 | 74±10 |
| MAP, mm Hg* | 92±12 | 91±12 | 95±12 |
| Hypertension treatment | 1958 (23.7) | 1176 (18.0) | 1561 (31.2) |
| History of MI | 171 (2.1) | 108 (1.7) | 128 (2.6) |
| History of CHF | 89 (1.1) | 42 (0.6) | 71 (1.4) |
| CFPWV, m/s | 7.7 [6.6, 9.6] | --- | --- |
| CPP, mm Hg | 51.6 [43.1, 62.9] | --- | --- |
| FMD, % | --- | 4.5 [2.0, 7.2] | --- |
| Hyperemic velocity, cm/s | --- | 57.7 [43.2, 71.2] | --- |
| Baseline mean flow, cm/s | --- | 6.5 [4.5, 9.4] | --- |
| PAT ratio | --- | --- | 2.0 [1.5, 2.7] |
| Incident VTE | 185 (2.2) | 130 (2.0) | 82 (1.6) |
| Type | |||
| DVT | 108 (58.4) | 77 (59.2) | 46 (56.1) |
| PE | 57 (30.8) | 38 (29.2) | 24 (29.3) |
| DVT and PE | 20 (10.8) | 15 (11.5) | 12 (14.6) |
| Etiology | |||
| Provoked | 125 (67.6) | 90 (69.2) | 53 (64.6) |
| Unprovoked | 52 (28.1) | 38 (29.2) | 27 (32.9) |
| Unknown | 125 (67.6) | 90 (69.2) | 53 (64.6) |
Abbreviations: CFPWV, carotid-femoral pulse wave velocity; CHF, congestive heart failure; CPP, central pulse pressure; DBP, diastolic blood pressure; FHS, Framingham Heart Study; FMD, flow-mediated dilation; MAP, mean arterial pressure; MI, myocardial infarction; PAT, peripheral arterial tonometry; SBP, systolic blood pressure
Note: Values in the table are N (%),mean±standard deviation, or median [25th, 75th percentile].
Mean arterial pressure is missing for 84 participants in sample #2 and 292 participants in sample #3.
Table 2.
Cox proportional hazards model results for the association between vascular function variables and incident VTE.
| Arterial Stiffness Measure (per SD*) | No. of events/No. of participants at risk | Model 1: Age/sex-adjusted | Model 2: Multivariable-adjusted** | ||
|---|---|---|---|---|---|
| P-value | P-value | ||||
| CFPWV*** | 185/8261 | ||||
| CPP | 185/8261 | ||||
| FMD | 130/6540 | ||||
| Hyperemic velocity | 130/6540 | ||||
| Ln-PAT ratio | 82/4998 | ||||
Abbreviations: CFPWV, carotid-femoral pulse wave velocity; CHF, congestive heart failure; CI, confidence interval; CPP, central pulse pressure; DBP, diastolic blood pressure; FHS, Framingham Heart Study; FMD, flow-mediated dilation; HR, hazard ratio; Ln, natural logarithm; MAP, mean arterial pressure; MI, myocardial infarction; PAT, peripheral arterial tonometry; SD, standard deviation; SBP, systolic blood pressure; VTE, venous thromboembolism
All models are stratified by cohort membership.
1 standard deviation = 35.4 ms/m for transformed-CFPWV, 17.1 mm Hg for CPP, 3.7 for FMD, 20.2 cm/s for hyperemic velocity, 0.40 for ln-PAT ratio
All models are adjusted heart rate, anti-hypertensive medication use, height, weight, diabetes, current smoking, history of myocardial infarction, history of heart failure, total/HDL cholesterol ratio, triglycerides, and lipid lowering medication use. The models for CFPWV and CPP are additionally adjusted for mean arterial pressure. The model for FMD is additionally adjusted for SBP, DBP, and hyperemic velocity. The model for hyperemic velocity is additionally adjusted for SBP, DBP, and baseline mean flow. The model for PAT ratio is additionally adjusted for SBP and DBP.
CFPWV was transformed by taking the inverse and multiplying by −1000 in order to convert units to ms/m and maintain the original directionality of the variable (higher value represents a stiffer aorta).
Table 3.
Cox proportional hazards model results for the association between vascular function variables and incident VTE, excluding prevalent cardiovascular disease at baseline.
| Arterial Stiffness Measure (per SD*) | No. of events/No. of participants | Model 1: Age/sex-adjusted | Model 2: Multivariable-adjusted** | ||
|---|---|---|---|---|---|
| HR (95% CI) | P-value | HR (95% CI) | P-value | ||
| CFPWV*** | 157/7577 | 1.31 (1.02–1.69) | 0.03 | 1.29 (0.97–1.72) | 0.08 |
| CPP | 157/7577 | 0.91 (0.77–1.08) | 0.28 | 0.87 (0.71–1.07) | 0.19 |
| FMD | 119/6156 | 0.76 (0.59–0.99) | 0.04 | 0.79 (0.60–1.04) | 0.09 |
| Hyperemic velocity | 119/6156 | 0.85 (0.69–1.05) | 0.12 | 0.84 (0.66–1.05) | 0.12 |
| Ln PAT ratio | 67/4515 | 0.93 (0.72–1.19) | 0.54 | 0.93 (0.72–1.21) | 0.59 |
Abbreviations: CFPWV, carotid-femoral pulse wave velocity; CI, confidence interval; CPP, central pulse pressure; DBP, diastolic blood pressure; FMD, flow-mediated dilation; HR, hazard ratio; Ln, natural logarithm; PAT, peripheral arterial tonometry; SBP, systolic blood pressure; VTE, venous thromboembolism
All models are stratified by cohort membership.
1 standard deviation = 35.4 ms/m for transformed-CFPWV, 17.1 mm Hg for CPP, 3.7 for FMD, 20.2 cm/s for hyperemic velocity, 0.40 for ln-PAT ratio.
All models are adjusted heart rate, anti-hypertensive medication use, height, weight, diabetes, current smoking, history of myocardial infarction, history of heart failure, total/HDL cholesterol ratio, triglycerides, and lipid lowering medication use. The models for CFPWV and CPP are additionally adjusted for mean arterial pressure. The model for FMD is additionally adjusted for SBP, DBP, and hyperemic velocity. The model for hyperemic velocity is additionally adjusted for SBP, DBP, and baseline mean flow. The model for PAT ratio is additionally adjusted for SBP and DBP.
CFPWV was transformed by taking the inverse and multiplying by −1000 in order to convert units to ms/m and maintain the original directionality of the variable (higher value represents a stiffer aorta).
Discussion
In our community-based prospective cohort study, the rate of incident VTE was ≈ 2 per 1000 person-years during a mean follow-up up to 11 years. We observed that CFPWV was associated with risk of incident VTE after age- and sex-adjustment, however, the association was no longer statistically significant after multivariable adjustment. None of the other measures of endothelial dysfunction were significantly associated with risk of VTE.
Although arterial and venous thromboses have been thought to be two distinct diseases, there is now emerging data suggesting that there may be common underlying mechanisms for both. Atherosclerosis and VTE share common cardiovascular risk factors,[17,19] and patients with VTE have been shown to have higher risk of subclinical atherosclerosis.[20] In a Mendelian randomization analysis restricted to Europeans, genetic predisposition to higher low-density lipoprotein cholesterol level was associated with higher risk of incident VTE.[21] The JUPITER trial (Justification for the Use of Statins in Prevention: An Intervention Trial Evaluating Rosuvastatin) demonstrated a significant reduction in VTE, both provoked and unprovoked, with rosuvastatin treatment.[22] Furthermore, in the combined analysis of FOURIER (Further Cardiovascular Outcomes Research With PCSK9 Inhibition in Subjects With Elevated Risk) and ODYSSEY OUTCOMES (Evaluation of Cardiovascular Outcomes After an Acute Coronary Syndrome During Treatment with Alirocumab) trials, PCSK9 was shown to reduce the risk of VTE compared to placebo, and the effect appeared to be mediated by reduction in Lp(a) level.[23]
Impaired endothelium-dependent vasodilation is a hallmark of atherosclerosis, and endothelium-dependent regulation of venous vasculature likely plays a key role in pathogenesis of VTE. There have been limited numbers of observational studies investigating associations between measures of endothelial dysfunction and the risk of VTE (Supplemental Table 2). The study by Prasad et al.[6] is the only prospective cohort study published to date, in which 54 patients with and 51 without coronary endothelial dysfunction, assessed with reaction to intracoronary acetylcholine, were followed for the median follow-up of 6.3 years. There were 6 incident VTE events, and they all occurred in patients with coronary endothelial dysfunction. Other observational studies[24–27] also reported an association between FMD or PAT ratio and risk of VTE in cross-sectional or case-control studies.
There are several limitations of the current study that may account for the lack of association between the measures of arterial function and VTE. Although our study is the largest prospective cohort study to examine the association between arterial vascular function and risk of VTE, our power to detect significant association may have been limited by the low number of VTE events. Future meta-analyses incorporating multiple prospective cohorts may elucidate the relation between arterial vascular function and risk of VTE further. Our study cohort was predominantly white and of European ancestry from New England, and our results may not be generalizable to other races or ethnicities or regions. The vascular function measurements and covariates were measured at a fixed time and were not updated over time. In addition, our study samples were drawn from different cohorts at different exams, and VTE ascertainment may have changed over time. Finally, we did not compare our results with established risk scores for VTE. Since we did not observe a significant association between the vascular function measurements and risk of incident VTE after multivariable adjustment, arterial vascular function is ujnlikely to contribute to risk prediction in VTE. Our study has several important strengths. In our community-based prospective cohort approach, participants were rigorously followed over two decades. All VTE events were validated by a panel of three physicians. Finally, we provide results of association between a comprehensive set of vascular function measures and VTE.
Our battery of vascular function measures was designed to provide a reasonably comprehensive assessment of large and small artery structure and function. CFPWV is the reference standard measure of arterial wall stiffness and is an important risk factor for cardiovascular disease, particularly in middle aged adults. [28,29] CPP is closely related to matching between wall stiffness and flow-diameter relations of the proximal aorta. Excessive wall stiffness or high flow for a given diameter increases CPP. FMD, hyperemic velocity and PAT ratio are measures of endothelial function and microvascular structure and function.
In conclusion, in our comprehensive examination of arterial vascular function and risk of VTE, we found CFPWV to be associated with the risk of incident VTE after adjustment for age and sex. We did not observe an association between any vascular function measure and incident VTE, after multivariable adjustment.
Supplementary Material
Highlights.
Association between measures of arterial vascular function and risk of incident venous thromboembolism (VTE) was examined in Framingham Heart Study
Carotid-femoral pulse wave velocity was associated with increased risk of incident VTE after adjusting for age and sex. The association was no longer statistically significant after multivariable adjustment.
Central pulse pressure, flow-mediated dilation, hyperemic velocity, and peripheral arterial tonometry were not associated with risk of incident VTE in both age- and sex-adjusted and multivariable-adjusted analyses.
Funding:
The Framingham Heart Study (FHS) acknowledges the support of Contracts NO1-HC-25195, HHSN268201500001I and 75N92019D00031 from the National Heart, Lung and Blood Institute (NHLBI) and grants HL107385, HL126136, HL93328, HL142983, HL143227 and HL131532 for this research. We also acknowledge the dedication of the FHS study participants without whom this research would not be possible. The views expressed in this manuscript are those of the authors and do not necessarily represent the views of the National Heart, Lung, and Blood Institute; the National Institutes of Health; or the U.S. Department of Health and Human Services. Dr. Ko is supported by American College of Cardiology Foundation/Merck Research Fellowship in Cardiovascular Diseases and Cardiometabolic Disorders. Dr. Johnson is supported by NHLBI Division of Intramural Research Funds. Dr. Vasan is supported in part by the Evans Medical Foundation and the Jay and Louis Coffman Endowment from the Department of Medicine, Boston University School of Medicine. Dr. Benjamin is supported by NHLBI R01HL092577, American Heart Association AF Strategically Focused Research Network 18SFRN34110082, and American Heart Association Tobacco Regulation and Addiction Center 2U54HL120163. Dr. Mitchell was funded by research grants HL094898, DK082447, HL104184, HL107385, HL126136 and HL142983 from the National Institutes of Health.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Disclosures
Dr. Mitchell is owner of Cardiovascular Engineering, Inc., a company that develops and manufactures devices to measure vascular stiffness, serves as a consultant to and receives honoraria from Novartis, Merck, Bayer, Servier and Philips.
Declaration of interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References
- 1.Heit JA. Epidemiology of venous thromboembolism. Nat Rev Cardiol. 2015;12:464–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Office of the Surgeon General (US) and National Heart L, and Blood Institute (US). The Surgeon General’s call to action to prevent deep vein thrombosis and pulmonary embolism. 2008. [PubMed]
- 3.Streiff MB, Brady JP, Grant AM, Grosse SD, Wong B and Popovic T. CDC Grand Rounds: preventing hospital-associated venous thromboembolism. MMWR Morb Mortal Wkly Rep. 2014;63:190–3. [PMC free article] [PubMed] [Google Scholar]
- 4.Mahmoodi BK, Cushman M, Anne Naess I, Allison MA, Bos WJ, Braekkan SK, Cannegieter SC, Gansevoort RT, Gona PN, Hammerstrom J, et al. Association of Traditional Cardiovascular Risk Factors With Venous Thromboembolism: An Individual Participant Data Meta-Analysis of Prospective Studies. Circulation. 2017;135:7–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Prandoni P, Bilora F, Marchiori A, Bernardi E, Petrobelli F, Lensing AW, Prins MH and Girolami A. An association between atherosclerosis and venous thrombosis. N Engl J Med. 2003;348:1435–41. [DOI] [PubMed] [Google Scholar]
- 6.Prasad M, McBane R, Reriani M, Lerman LO and Lerman A. Coronary endothelial dysfunction is associated with increased risk of venous thromboembolism. Thromb Res. 2016;139:17–21. [DOI] [PubMed] [Google Scholar]
- 7.Aykan A, Hatem E, Kalaycıoğlu E, Gökdeniz T and Karabay CY. Assessment of arterial stiffness in patients with venous thromboembolism: Separate or continuous circuits? Phlebology. 2017;32:316–321. [DOI] [PubMed] [Google Scholar]
- 8.Duran NE, Oğuz E, Duran I, Aykan AC, Ertürk E, Bayram Z, Gürsoy MO, Biteker M, Karavelioğlu Y, Kaya H, et al. Aortic elastic properties in patients with venous thromboembolism. Phlebology. 2010;25:246–51. [DOI] [PubMed] [Google Scholar]
- 9.Prandoni P. Venous thromboembolism and atherosclerosis: is there a link? J Thromb Haemost. 2007;5 Suppl 1:270–5. [DOI] [PubMed] [Google Scholar]
- 10.Holst AG, Jensen G and Prescott E. Risk factors for venous thromboembolism: results from the Copenhagen City Heart Study. Circulation. 2010;121:1896–903. [DOI] [PubMed] [Google Scholar]
- 11.Ageno W, Becattini C, Brighton T, Selby R and Kamphuisen PW. Cardiovascular risk factors and venous thromboembolism: a meta-analysis. Circulation. 2008;117:93–102. [DOI] [PubMed] [Google Scholar]
- 12.Tsao CW and Vasan RS. Cohort Profile: The Framingham Heart Study (FHS): overview of milestones in cardiovascular epidemiology. Int J Epidemiol. 2015;44:1800–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Benjamin EJ, Larson MG, Keyes MJ, Mitchell GF, Vasan RS, Keaney JF Jr., Lehman BT, Fan S, Osypiuk E and Vita JA. Clinical correlates and heritability of flow-mediated dilation in the community: the Framingham Heart Study. Circulation. 2004;109:613–9. [DOI] [PubMed] [Google Scholar]
- 14.Mitchell GF, Parise H, Vita JA, Larson MG, Warner E, Keaney JF Jr., Keyes MJ, Levy D, Vasan RS and Benjamin EJ. Local shear stress and brachial artery flow-mediated dilation: the Framingham Heart Study. Hypertension. 2004;44:134–9. [DOI] [PubMed] [Google Scholar]
- 15.Hamburg NM, Palmisano J, Larson MG, Sullivan LM, Lehman BT, Vasan RS, Levy D, Mitchell GF, Vita JA and Benjamin EJ. Relation of brachial and digital measures of vascular function in the community: the Framingham heart study. Hypertension. 2011;57:390–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Hamburg NM, Keyes MJ, Larson MG, Vasan RS, Schnabel R, Pryde MM, Mitchell GF, Sheffy J, Vita JA and Benjamin EJ. Cross-sectional relations of digital vascular function to cardiovascular risk factors in the Framingham Heart Study. Circulation. 2008;117:2467–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Puurunen MK, Gona PN, Larson MG, Murabito JM, Magnani JW and O’Donnell CJ. Epidemiology of venous thromboembolism in the Framingham Heart Study. Thromb Res. 2016;145:27–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Lin DY, Wei LJ and Ying Z. Checking the Cox Model with Cumulative Sums of Martingale-Based Residuals. Biometrika. 1993;80:557–572. [Google Scholar]
- 19.Dentali F, Gessi V, Marcucci R, Gianni M, Grandi AM and Franchini M. Lipoprotein(a) as a Risk Factor for Venous Thromboembolism: A Systematic Review and Meta-analysis of the Literature. Semin Thromb Hemost. 2017;43:614–620. [DOI] [PubMed] [Google Scholar]
- 20.Prandoni P, Bilora F, Marchiori A, Bernardi E, Petrobelli F, Lensing AWA, Prins MH and Girolami A. An Association between Atherosclerosis and Venous Thrombosis. N Engl J Med. 2003;348:1435–1441. [DOI] [PubMed] [Google Scholar]
- 21.Klarin D, Busenkell E, Judy R, Lynch J, Levin M, Haessler J, Aragam K, Chaffin M, Haas M, Lindström S, et al. Genome-wide association analysis of venous thromboembolism identifies new risk loci and genetic overlap with arterial vascular disease. Nat Genet. 2019;51:1574–1579. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Glynn RJ, Danielson E, Fonseca FA, Genest J, Gotto AM Jr, JJ Kastelein, Koenig W, Libby P, AJ Lorenzatti, JG MacFadyen, et al. A randomized trial of rosuvastatin in the prevention of venous thromboembolism. N Engl J Med. 2009;360:1851–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Marston NA, Gurmu Y, Melloni GEM, Bonaca M, Gencer B, Sever PS, Pedersen TR, Keech AC, Roselli C, Lubitz SA, et al. The Effect of PCSK9 (Proprotein Convertase Subtilisin/Kexin Type 9) Inhibition on the Risk of Venous Thromboembolism. Circulation. 2020;141:1600–1607. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Migliacci R, Becattini C, Pesavento R, Davi G, Vedovati MC, Guglielmini G, Falcinelli E, Ciabattoni G, Dalla Valle F, Prandoni P, et al. Endothelial dysfunction in patients with spontaneous venous thromboembolism. Haematologica. 2007;92:812–8. [DOI] [PubMed] [Google Scholar]
- 25.Mazzoccoli G, Fontana A, Grilli M, Dagostino MP, Copetti M, Pellegrini F and Vendemiale G. Idiopathic deep venous thrombosis and arterial endothelial dysfunction in the elderly. Age (Dordr). 2012;34:751–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Suzuki H, Matsuzawa Y, Konishi M, Akiyama E, Takano K, Nakayama N, Kataoka S, Ebina T, Kosuge M, Hibi K, et al. Utility of noninvasive endothelial function test for prediction of deep vein thrombosis after total hip or knee arthroplasty. Circ J. 2014;78:1723–32. [DOI] [PubMed] [Google Scholar]
- 27.Jeraj L, Spirkoska A, Ježovnik MK and Poredoš P. Deep vein thrombosis and properties of the arterial wall. Vasa. 2018;47:197–201. [DOI] [PubMed] [Google Scholar]
- 28.Mitchell GF, Hwang SJ, Vasan RS, Larson MG, Pencina MJ, Hamburg NM, Vita JA, Levy D and Benjamin EJ. Arterial stiffness and cardiovascular events: the Framingham Heart Study. Circulation. 2010;121:505–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Ben-Shlomo Y, Spears M, Boustred C, May M, Anderson SG, Benjamin EJ, Boutouyrie P, Cameron J, Chen CH, Cruickshank JK, et al. Aortic pulse wave velocity improves cardiovascular event prediction: an individual participant meta-analysis of prospective observational data from 17,635 subjects. J Am Coll Cardiol. 2014;63:636–646. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.