Abstract
Background
Pulse wave velocity (PWV) is a noninvasive measure of arterial stiffness and predictor of cardiovascular disease. However, the association between PWV and vascular calcification across different vascular beds has not been fully investigated. This study aimed to quantify the association between PWV and multiterritory calcification and to explore whether PWV can identify individuals with vascular calcification beyond traditional risk factors.
Methods and Results
Among 1351 older adults (mean age, 79.2 years [SD, 4.1]) from the ARIC (Atherosclerosis Risk in Communities) study, we measured segment‐specific PWVs: heart–carotid, heart–femoral, carotid–femoral, heart–ankle, brachial–ankle, and femoral–ankle. Dependent variables were high calcium score (≥75th percentile of Agatston score) across different vascular beds: coronary arteries, aortic valve ring, aortic valve, mitral valve, ascending aorta, and descending aorta. Quartiles of carotid–femoral, heart–femoral, heart–ankle, and brachial–ankle PWV were significantly associated with coronary artery calcium (eg, adjusted odds ratio [OR] for the highest versus lowest quartile of carotid–femoral PWV, 1.84 [95% CI, 1.24–2.74]). Overall, PWVs were most strongly associated with descending aorta calcification, with significant results for carotid–femoral, heart–femoral, heart–ankle, and brachial–ankle PWV (eg, adjusted OR for the highest versus lowest quartile of carotid–femoral PWV, 3.99 [95% CI, 2.61–6.17]). In contrast, femoral–ankle PWV was inversely associated with descending aorta calcification. Some PWVs improved the discrimination of coronary artery calcium and descending aorta calcification beyond traditional risk factors.
Conclusions
The associations of PWV with vascular calcification varied substantially across segments, with descending aorta calcification most closely linked to PWVs. Our study suggests that some PWVs, especially carotid–femoral PWV, are helpful to identify individuals with coronary artery calcium and descending aorta calcification.
Keywords: aortic calcification, arterial stiffness, computed tomography, coronary artery calcium, pulse wave velocity, valvular calcification, vascular calcification
Subject Categories: Cardiovascular Disease, Epidemiology, Risk Factors, Computerized Tomography (CT)
Nonstandard Abbreviations and Acronyms
- ARIC
Atherosclerosis Risk in Communities
- baPWV
brachial–ankle pulse wave velocity
- cfPWV
carotid–femoral pulse wave velocity
- ECC
extracoronary calcium
- faPWV
femoral–ankle pulse wave velocity
- haPWV
heart–ankle pulse wave velocity
- hfPWV
heart–femoral pulse wave velocity
- PWV
pulse wave velocity
Clinical Perspective.
What Is New?
Pulse wave velocity (PWV) is a noninvasive measure of arterial stiffness and predictor of cardiovascular disease; however, the associations between PWV and vascular calcification across different vascular beds have not been fully investigated.
In 1351 older adults from a US community‐based cohort, several PWV measures, particularly carotid–femoral PWV, were independently associated with coronary artery calcification and calcification of other vascular beds, such as descending aorta calcification.
Among different vascular beds, descending aorta calcification was most strongly associated with PWV measures.
What Are the Clinical Implications?
Our study further supports a unique pathophysiological process across different vascular beds and suggests that some PWV measures, especially carotid–femoral PWV, could help identify individuals with coronary artery calcium or descending aorta calcification.
Arterial stiffness is a pathophysiological process contributing to the development of cardiovascular disease (CVD) and end‐organ damage. 1 Among various measures of arterial stiffness, pulse wave velocity (PWV) is most widely used in clinical and research settings. 2 Among PWV measures, carotid–femoral PWV (cfPWV) is considered as a gold standard for the evaluation of arterial stiffness 3 and has been shown to predict CVD risk beyond traditional risk factors. 4 Furthermore, PWV measurement is noninvasive, radiation free, and relatively easy for appropriately trained personnel. 2
Vascular calcification is another arterial manifestation with prognostic value. Coronary artery calcium (CAC) has been one of the strongest predictors of CVD. 5 The American College of Cardiology and American Heart Association guidelines recommend CAC for guiding CVD preventive therapy (eg, statin) in individuals with intermediate risk based on traditional risk classification. 6 Similarly, extracoronary calcification (ECC) (eg, calcification of the aorta or mitral valve) is also independently associated with CVD. 7 , 8 , 9 , 10 , 11 However, the measurement of CAC and ECC is usually done by computed tomography (CT) and thus requires radiation in an imaging center.
Reflecting PWV and vascular calcification as measures of arterial wall pathophysiology, several studies have reported positive associations between PWV and CAC. 12 , 13 , 14 , 15 , 16 However, there are several important knowledge gaps. Most of these studies focused on cfPWV or brachial–ankle PWV (baPWV), and associations of other PWV measures (eg, heart–femoral PWV [hfPWV]) with CAC are uncertain. Few studies have explored ECC and its association with PWV measures. Additionally, PWV may be helpful to identify people with vascular calcification who may benefit from a formal assessment of vascular calcification with CT, but to the best of our knowledge, no studies have formally evaluated whether PWV measures improve the discrimination of having vascular calcification beyond traditional risk factors.
Therefore, the present study aimed to (1) evaluate associations between different PWV measures and different vascular calcification including CAC and (2) assess whether PWV measures improve discrimination of CAC and calcification of other vascular beds beyond traditional risk factors using a community‐based cohort, the ARIC (Atherosclerosis Risk in Communities) study.
METHODS
Data of the ARIC study are available through the National Heart, Lung, and Blood Institute Biologic Specimen and Data Repository Information Coordinating Center (https://biolincc.nhlbi.nih.gov/studies/aric/) or from the ARIC study coordinating center at University of North Carolina, Chapel Hill, North Carolina.
Study Population
Details of the ARIC study have been published elsewhere. 17 The ARIC study is a prospective cohort study including 15 792 participants aged 45 to 64 years at visit 1 (1987–1989) from 4 US communities (Forsyth County, NC; Jackson, MS; suburban Minneapolis, MN; and Washington County, MD). The follow‐up examination to date included visits 2 (1990–1992), 3 (1993–1995), 4 (1996–1998), 5 (2011–2013), 6 (2016–2017), 7 (2018–2019), 8 (2020), and 9 (2021–2022). ARIC was approved by the institutional review boards of the participating centers and was conducted according to the principles expressed in the Declaration of Helsinki. All participants provided written informed consent for attending this study.
Electrocardiogram‐gated CT scans to assess CAC and ECC were first conducted at visit 7, whereas participants were invited for PWV assessment across visits 6 and 7. Of the 3589 participants at visit 7, we excluded participants who did not undergo CT scan due to clinical history of myocardial infarction (adjudicated definite or probable cases) or coronary revascularization (based on International Classification of Diseases [ICD] Ninth or Tenth R e v is io n, as appropriate) (n=1271), those who did not undergo PWV measures either at visits 6 or 7 (n=639), those with missing value in any PWV measures of interest as described below (n=232), those with a missing value in covariates (n=90), and Asian or Native American participants (n=6) due to a small sample compared with Blacks and Whites. A total of 1351 participants were included in the present study (Figure S1). Of these, 784 participants and 567 participants completed PWV evaluation at visits 6 and 7, respectively.
Pulse Wave Velocity
PWV was measured using the VP‐1000plus (Omron Healthcare). 18 We analyzed the following PWV measures across different vascular segments in the present study: cfPWV; heart–carotid PWV; hfPWV; heart–ankle (haPWV); baPWV, and femoral–ankle (faPWV). We recorded carotid and femoral arterial pressure waveforms using applanation tonometry sensors at the left common carotid artery and left common femoral artery. Bilateral brachial and posttibial arterial pressure waveforms were also recorded for 10 seconds with cuffs wrapped on both arms and ankles. The details of PWV calculation are in Data S1. Based on anatomy, cfPWV, heart–carotid PWV, and hfPWV are considered to reflect central arterial stiffness, whereas faPWV represents peripheral artery stiffness. 19 Moreover, baPWV and haPWV reflect both central and peripheral arterial stiffness. 20 For baPWV, haPWV, and faPWV, we used the higher value of their bilateral data. All PWV measures were taken at least twice and averaged.
Vascular Calcification
Calcium score was measured by noncontrast, cardiac‐gated CT and was calculated using the Agatston method (Agatston score). 21 Calcification was defined as lesions with attenuation ≥130 Hounsfield units (HU) and area ≥1 mm2 in each slice level. Following the standard approach to obtain the Agatston score, for each calcified lesion, the lesion score was obtained by multiplying the lesion area and the density factor (1 if 130–199 HU, 2 if 200–299 HU, 3 if 300–399 HU, and 4 if ≥400 HU). Then, the total Agatston score was calculated as the sum of all lesion scores. 22 The present study analyzed CAC and other vascular segments or valves as ECC: aortic valve ring, aortic valve, mitral valve, ascending aorta, and descending aorta. The images of coronary arteries and extracoronary arteries were acquired from the level of the pulmonary artery bifurcation to the apex of the heart; thus, aortic arch was not included. Aortic valve calcification was defined as calcification on the aortic valve extending to but not including the aortic root. Aortic valve ring calcification indicates calcification of aortic wall above aortic valve but below coronary arteries. Ascending aorta and descending aorta calcification were measured within the image ranges (ie, aortic arch was not included) (Figure S2).
Covariates
We used covariates from visits 6 or 7 unless specified otherwise, aligning with the timing of the PWV measurement. Educational level was assessed at visit 1 and categorized into basic (less than high school), intermediate (graduated from high school), and advanced (at least some college). Body mass index was defined as weight (in kilograms) divided by height (in meters) squared. Antihypertensive medication use was defined by self‐report usage in the past 4 weeks and inspection of medication bottles brought to the visit. Blood pressure was measured by certified technicians with participants in the seated position after a 5‐minute rest, and the average of the last 2 of 3 readings was recorded. History of diabetes was based on fasting plasma glucose ≥126 mg/dL, nonfasting glucose ≥200 mg/dL, using medication for diabetes, or self‐report of diagnosis by physician. Total and high‐density lipoprotein cholesterol were measured using the Olympus high‐density lipoprotein cholesterol test. Estimated glomerular filtration rate was computed using the Chronic Kidney Disease Epidemiology Collaboration 2021 creatinine and cystatin C equation. 23 Prior coronary heart disease was defined as self‐reported clinical history at visit 1 and then adjudicated events by a physician panel between visit 1 and the relevant visit. Similarly, prior stroke was defined as a self‐reported clinical history at visit 1 or adjudicated events during the study follow‐up. Prior heart failure was defined as prior hospitalization adjudicated by a physician panel (from 2005 to the relevant visit) or hospitalization with heart failure by ICD‐based diagnosis (ICD‐9 code 428.X or ICD‐10 code I‐50) before 2005. Smoking status and drinking alcohol were categorized as current and noncurrent. Because leg artery stenosis and heart rate may affect PWV measures, we further adjusted for ankle–brachial index and heart rate as covariates as sensitivity analyses. The lower ankle–brachial index measure of its bilateral data was used for the analysis, which was simultaneously measured using the same device when PWV was assessed. Heart rate was also simultaneously measured by electrocardiography during PWV measurement.
Statistical Analysis
Because we were interested in PWV measures as predictors of CAC, we treated PWV measures as independent (exposure) variables and CAC/ECC as dependent (outcome) variables. We showed participant characteristics across quartiles of PWV measures. Continuous variables were summarized as mean value with SD, and categorical variables were described as count with proportion.
To assess the association of each PWV measure with CAC and ECC, we ran logistic regression models, with the lowest quartile of each PWV measure as a reference and estimated odds ratio (OR) and 95% CI. We defined CAC and ECC ≥75th percentile Agatston score as high calcification in this analysis. We constructed 3 models to assess the impact of potential confounders: Model 1 unadjusted; Model 2 adjusting for age, sex, race, and field center; and Model 3 further accounting for education level, body mass index, antihypertensive medication use, systolic blood pressure, diabetes, total and high‐density lipoprotein cholesterol levels, estimated glomerular filtration rate, prior coronary heart disease, prior heart failure, prior stroke, current drinker, and current smoker. We also modeled each PWV as a continuous variable with restricted cubic spline terms and covariates in Model 3 noted above. We set the median in the lowest quartile as a reference and put 4 knots at 5th, 35th, 65th, and 95th percentiles of each PWV measure.
As sensitivity analysis, we first conducted subgroup analysis using Model 3 by age (<78 versus ≥78 years [median]), sex (men versus women), race (Black versus White), and systolic blood pressure (<133 versus ≥133 mm Hg [median]). To obtain reliable estimates in each subgroup, we modeled relevant PWV measures as a linear term (per 1 SD) and tested interaction using likelihood ratio tests. As noted above, we further adjusted for ankle–brachial index or heart rate as Models 4 and 5. We also excluded subpopulations with conditions influencing PWV measurement such as severe obesity (body mass index ≥40 kg/m2) (n=11) or peripheral revascularization history before PWV measures (n=7), possible outliers (PWV ≥3 SD above the mean; n=61), or aortic stenosis (n=27), defined as aortic valve peak velocity >2.0 m/s by echocardiography or incomplete assessment (n=25).
Finally, we evaluated whether adding PWV measures improves discrimination of high CAC and ECC beyond traditional CVD risk factors. The base model included predictors in Model 3 described above. We assessed C statistics as a measure of discrimination using logistic regression models. The C statistic between base model and adding PWV measures was compared using the DeLong test. We also assessed Akaike information criterion and Bayesian information criterion for comparing the model performance after adding each PWV measure. Statistical significance was set at P<0.05 (2‐sided). All analyses were performed using R version 4.2.0 (The R Foundation for Statistical Computing, Vienna, Austria).
RESULTS
Baseline Characteristics
Among 1351 participants, the mean age was 79.2 years (SD, 4.1 years), 39% were men, 19% were Black, 27% had diabetes, and 71% were taking antihypertensive medication. Participants with higher cfPWV tended to have higher risk profiles (Table 1). Generally, similar patterns were observed across quartiles of other PWV measures, although heart–carotid PWV and faPWV demonstrated somewhat unique characteristics (Tables S1 through S5).
Table 1.
Baseline Characteristics According to Quartiles of cfPWV
| Variables | Overall n=1351 | Quartile 1 n=337 | Quartile 2 n=338 | Quartile 3 n=338 | Quartile 4 n=338 |
|---|---|---|---|---|---|
| PWV range, cm/s | 353–6155 | 353–1011.5 | 1012–1196 | 1196.5–1414.5 | 1415.5–6155 |
| Age, y | 79.2 (4.1) | 78.7 (4.0) | 78.4 (3.9) | 79.5 (4.0) | 80.3 (4.4) |
| Men | 527 (39.0) | 119 (35.3) | 129 (38.2) | 137 (40.5) | 142 (42.0) |
| Black | 250 (18.5) | 49 (14.5) | 62 (18.3) | 58 (17.2) | 81 (24.0) |
| Body mass index, kg/m2 | 27.6 (4.5) | 27.7 (4.6) | 27.9 (4.8) | 27.7 (4.1) | 27 (4.4) |
| Education level | |||||
| Basic | 115 (8.5) | 20 (5.9) | 23 (6.8) | 37 (10.9) | 35 (10.4) |
| Intermediate | 565 (41.8) | 133 (39.5) | 141 (41.7) | 133 (39.3) | 158 (46.7) |
| Advanced | 671 (49.7) | 184 (54.6) | 174 (51.5) | 168 (49.7) | 145 (42.9) |
| Current drinker | 780 (57.7) | 215 (63.8) | 205 (60.7) | 201 (59.5) | 159 (47.0) |
| Current smoker | 81 (6.0) | 25 (7.4) | 18 (5.3) | 19 (5.6) | 19 (5.6) |
| Antihypertensive medication | 959 (71.0) | 229 (68.0) | 226 (66.9) | 248 (73.4) | 256 (75.7) |
| Systolic blood pressure, mm Hg | 134 (18) | 128 (17) | 131 (16) | 137 (18) | 141 (18) |
| Diastolic blood pressure, mm Hg | 67 (11) | 67 (11) | 67 (10) | 68 (11) | 68 (11) |
| Heart rate, bpm | 63 (10) | 61 (9) | 62 (10) | 64 (10) | 66 (11) |
| Total cholesterol, mmol/L | 4.6 (1.0) | 4.7 (1.0) | 4.6 (1.0) | 4.6 (1.1) | 4.6 (1.0) |
| HDL cholesterol, mmol/L | 1.4 (0.4) | 1.4 (0.4) | 1.4 (0.3) | 1.3 (0.4) | 1.3 (0.4) |
| eGFR, mL/min per 1.73 m2 | 67.3 (17.5) | 69.4 (17.9) | 69 (16.8) | 65.9 (17.3) | 64.9 (17.6) |
| Diabetes | 361 (26.7) | 58 (17.2) | 74 (21.9) | 103 (30.5) | 126 (37.3) |
| Prior coronary heart disease | 27 (2.0) | 3 (0.9) | 5 (1.5) | 7 (2.1) | 12 (3.6) |
| Prior heart failure | 36 (2.7) | 7 (2.1) | 9 (2.7) | 10 (3.0) | 10 (3.0) |
| Prior stroke | 38 (2.8) | 8 (2.4) | 7 (2.1) | 8 (2.4) | 15 (4.4) |
| Ankle–brachial index | 1.10 (0.12) | 1.09 (0.11) | 1.12 (0.12) | 1.10 (0.12) | 1.08 (0.14) |
| Field center | |||||
| Forsyth County, NC | 270 (20.0) | 45 (13.4) | 61 (18.0) | 80 (23.7) | 84 (24.9) |
| Jackson, MS | 236 (17.5) | 45 (13.4) | 58 (17.2) | 57 (16.9) | 76 (22.5) |
| Minneapolis, MN | 500 (37.0) | 162 (48.1) | 131 (38.8) | 119 (35.2) | 88 (26.0) |
| Washington County, MD | 345 (25.5) | 85 (25.2) | 88 (26.0) | 82 (24.3) | 90 (26.6) |
Continuous variables with mean (SD) and categorical variables with number (percent) are presented. cfPWV indicates carotid–femoral pulse wave velocity; eGFR, estimated glomerular filtration rate; and HDL, high‐density lipoprotein.
PWV and Vascular Calcification
We observed heterogeneous associations across different PWV measures and the calcification of different vascular beds (Figure 1). For CAC, cfPWV demonstrated the strongest association, followed by hfPWV and haPWV (Figure 1A). Among different vascular beds, descending aorta calcification demonstrated the strongest association with multiple PWV measures (ie, hfPWV, cfPWV, haPWV, and baPWV) (Figure 1F), followed by CAC (Figure 1A) and mitral valve (Figure 1D). Aortic valve ring, aortic valve, and ascending aorta calcifications were not associated with any PWV measures in this analysis. From the perspective of PWV measures, overall, cfPWV (green dots in Figure 1) showed the most robust associations, with significant OR for CAC, mitral valve calcification, and descending aorta calcification. faPWV (pink dots in Figure 1) was inversely associated with vascular calcification, although statistical significance was seen only for descending aorta calcification. Generally, similar patterns were seen across our 3 models of covariates for CAC (Table 2) and descending aorta (Table 3). For calcification of the other vascular beds, their associations with PWV measures were mostly attenuated after accounting for potential confounders (Tables S6 through S9).
Figure 1. Adjusted odds ratio for different vascular calcification according to different PWV measures (the highest quartile vs the lowest quartile).
Each panel shows odds ratio for calcification of different arteries (A, coronary artery; B, aortic valve ring; C, aortic valve; D, mitral valve; E, ascending aorta; and F, descending aorta) according to different PWV measures (hc‐, hf‐, cf‐, ha‐, ba‐, and faPWV: the highest vs lowest quartile as reference). Odds ratio was estimated using multivariable logistic regression models adjusted for age, sex, race, field center, education level, body mass index, antihypertensive medication use, systolic blood pressure, diabetes, total and high‐density lipoprotein cholesterol levels, estimated glomerular filtration rate, prior coronary heart disease, prior heart failure, prior stroke, current drinker, and current smoker (n=1351). ba indicates brachial–ankle; cf, carotid–femoral; fa, femoral–ankle; ha, heart–ankle; hc, heart–carotid; hf, heart–femoral; and PWV, pulse wave velocity.
Table 2.
Odds Ratios for High Coronary Artery Calcium Score (≥75th Percentile) According to Quartiles of PWV Measures
| PWV measures | Median calcium score (IQI) | Prevalence/participants | Odds ratio (95% CI) | ||
|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | |||
| hcPWV, cm/s | |||||
| Quartile 1, 367–1006 | 208 (41–629) | 81/336 | 1 (reference) | 1 (reference) | 1 (reference) |
| Quartile 2, 1007.5–1204 | 172 (30–636) | 76/338 | 0.91 (0.64–1.31) | 0.88 (0.61–1.28) | 0.87 (0.59–1.28) |
| Quartile 3, 1205–1460 | 194 (38–666) | 85/339 | 1.05 (0.74–1.50) | 0.90 (0.62–1.31) | 0.94 (0.64–1.39) |
| Quartile 4, 1461.5–4994 | 273 (30–766) | 96/338 | 1.25 (0.89–1.76) | 0.90 (0.62–1.30) | 0.99 (0.67–1.45) |
| hfPWV, cm/s | |||||
| Quartile 1, 489–1086 | 154 (24–531) | 68/337 | 1 (reference) | 1 (reference) | 1 (reference) |
| Quartile 2, 1087.5–1230 | 163 (26–545) | 66/337 | 0.96 (0.66–1.41) | 0.98 (0.66–1.46) | 0.98 (0.66–1.48) |
| Quartile 3, 1231–1403 | 240 (46–767) | 94/339 | 1.52 (1.06–2.17)* | 1.36 (0.94–2.00) | 1.39 (0.94–2.06) |
| Quartile 4, 1403.5–2573.5 | 356 (66–870) | 110/338 | 1.91 (1.35–2.72)* | 1.68 (1.15–2.45)* | 1.65 (1.11–2.48)* |
| cfPWV, cm/s | |||||
| Quartile 1, 353–1011.5 | 144 (23–514) | 70/337 | 1 (reference) | 1 (reference) | 1 (reference) |
| Quartile 2, 1012–1196 | 143 (16–501) | 62/338 | 0.86 (0.58–1.25) | 0.82 (0.55–1.22) | 0.79 (0.52–1.19) |
| Quartile 3, 1196.5–1414.5 | 210 (42–660) | 84/338 | 1.26 (0.88–1.81) | 1.13 (0.77–1.66) | 1.05 (0.71–1.56) |
| Quartile 4, 1415.5–6155 | 403 (80–920) | 122/338 | 2.15 (1.53–3.05)* | 1.99 (1.38–2.89)* | 1.84 (1.24–2.74)* |
| haPWV, cm/s | |||||
| Quartile 1, 608.5–1068.5 | 148 (24–535) | 69/336 | 1 (reference) | 1 (reference) | 1 (reference) |
| Quartile 2, 1069–1160.5 | 210 (37–650) | 82/339 | 1.23 (0.86–1.78) | 1.18 (0.81–1.73) | 1.21 (0.82–1.79) |
| Quartile 3, 1160.7–1259 | 206 (33–648) | 81/337 | 1.22 (0.85–1.76) | 1.07 (0.73–1.57) | 1.17 (0.78–1.75) |
| Quartile 4, 1259.5–2064.5 | 303 (47–777) | 106/339 | 1.76 (1.24–2.51)* | 1.36 (0.93–1.98) | 1.55 (1.02–2.38)* |
| baPWV, cm/s | |||||
| Quartile 1, 612–1628.5 | 142 (23–592) | 74/335 | 1 (reference) | 1 (reference) | 1 (reference) |
| Quartile 2, 1629.5–1825.5 | 213 (39–614) | 76/340 | 1.02 (0.71–1.46) | 0.98 (0.67–1.44) | 0.98 (0.66–1.45) |
| Quartile 3, 1826–2063.3 | 192 (30–697) | 92/337 | 1.32 (0.93–1.89) | 1.23 (0.85–1.79) | 1.28 (0.86–1.91) |
| Quartile 4, 2064–3908 | 316 (46–770) | 96/339 | 1.39 (0.98–1.98) | 1.30 (0.89–1.89) | 1.27 (0.84–1.93) |
| faPWV, cm/s | |||||
| Quartile 1, 595–987.5 | 286 (41–780) | 100/337 | 1 (reference) | 1 (reference) | 1 (reference) |
| Quartile 2, 988.5–1092 | 185 (41–577) | 72/338 | 0.64 (0.45–0.91)* | 0.57 (0.40–0.83)* | 0.62 (0.42–0.90)* |
| Quartile 3, 1092.5–1224.5 | 165 (27–626) | 76/338 | 0.69 (0.49–0.97)* | 0.59 (0.41–0.85)* | 0.65 (0.44–0.95)* |
| Quartile 4, 1225–5748.5 | 228 (32–697) | 90/338 | 0.86 (0.61–1.20) | 0.65 (0.45–0.93)* | 0.70 (0.47–1.05) |
Model 1 was an unadjusted model. Model 2 was adjusted for covariates: age, sex, race, and field center. Model 3 was further adjusted for covariates other than Model 2: education level, body mass index, antihypertensive medication use, systolic blood pressure, diabetes, total and high‐density lipoprotein cholesterol level, estimated glomerular filtration rate, prior coronary heart disease, prior heart failure, prior stroke, current drinker, and current smoker. baPWV indicates brachial–ankle PWV; cfPWV, carotid–femoral PWV; faPWV, femoral–ankle PWV; haPWV, heart–ankle PWV; hcPWV, heart–carotid PWV; hfPWV, heart–femoral PWV; IQI, interquartile interval; and PWV, pulse wave velocity.
Indicate statistical significance.
Table 3.
Odds Ratios for Descending Aorta Calcification (≥75th Percentile) According to Quartiles of PWV Measures
| PWV measures | Median calcium score (IQI) | Prevalence/participants | Odds ratio (95% CI) | ||
|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | |||
| hcPWV, cm/s | |||||
| Quartile 1, 367–1006 | 415 (51–1431) | 85/336 | 1 (reference) | 1 (reference) | 1 (reference) |
| Quartile 2, 1007.5–1204 | 386 (38–1488) | 92/338 | 1.12 (0.80–1.58) | 1.16 (0.81–1.67) | 1.21 (0.83–1.78) |
| Quartile 3, 1205–1460 | 250 (28–1032) | 73/338 | 0.81 (0.57–1.16) | 0.90 (0.62–1.32) | 0.98 (0.66–1.45) |
| Quartile 4, 1461.5–4994 | 346 (39–1446) | 87/337 | 1.02 (0.72–1.45) | 1.05 (0.72–1.53) | 1.21 (0.81–1.81) |
| hfPWV, cm/s | |||||
| Quartile 1, 489–1086 | 138 (11–682) | 47/337 | 1 (reference) | 1 (reference) | 1 (reference) |
| Quartile 2, 1087.5–1230 | 287 (22–905) | 63/336 | 1.45 (0.96–2.19) | 1.50 (0.98–2.30) | 1.48 (0.95–2.32) |
| Quartile 3, 1231–1403 | 475 (55–1777) | 109/339 | 2.92 (2.00–4.32)* | 3.04 (2.05–4.56)* | 3.02 (1.98–4.65)* |
| Quartile 4, 1403.5–2573.5 | 649 (113–1965) | 118/337 | 3.31 (2.28–4.88)* | 3.42 (2.30–5.16)* | 3.13 (2.02–4.89)* |
| cfPWV, cm/s | |||||
| Quartile 1, 353–1011.5 | 107 (6–682) | 48/337 | 1 (reference) | 1 (reference) | 1 (reference) |
| Quartile 2, 1012–1196 | 232 (21–796) | 47/337 | 1.00 (0.65–1.54) | 1.04 (0.67–1.62) | 1.01 (0.64–1.60) |
| Quartile 3, 1196.5–1414.5 | 438 (69–1584) | 96/337 | 2.39 (1.63–3.54)* | 2.32 (1.56–3.48)* | 2.02 (1.33–3.10)* |
| Quartile 4, 1415.5–6155 | 927 (221–2633) | 146/338 | 4.58 (3.17–6.70)* | 4.56 (3.09–6.82)* | 3.99 (2.61–6.17)* |
| haPWV, cm/s | |||||
| Quartile 1, 608.5–1068.5 | 176 (13–860) | 63/336 | 1 (reference) | 1 (reference) | 1 (reference) |
| Quartile 2, 1069–1160.5 | 314 (35–1336) | 83/338 | 1.43 (0.99–2.07) | 1.41 (0.96–2.08) | 1.43 (0.95–2.15) |
| Quartile 3, 1160.7–1259 | 460 (42–1410) | 85/338 | 1.46 (1.01–2.12)* | 1.41 (0.96–2.07) | 1.37 (0.90–2.08) |
| Quartile 4, 1259.5–2064.5 | 522 (80–1705) | 106/337 | 1.97 (1.38–2.83)* | 1.68 (1.15–2.47)* | 1.66 (1.07–2.58)* |
| baPWV, cm/s | |||||
| Quartile 1, 612–1628.5 | 121 (6–741) | 49/335 | 1 (reference) | 1 (reference) | 1 (reference) |
| Quartile 2, 1629.5–1825.5 | 294 (32–1131) | 75/338 | 1.68 (1.13–2.51)* | 1.59 (1.06–2.39)* | 1.54 (1.01–2.37)* |
| Quartile 3, 1826–2063.3 | 409 (36–1474) | 88/339 | 2.06 (1.40–3.06)* | 1.78 (1.20–2.68)* | 1.63 (1.06–2.52)* |
| Quartile 4, 2064–3908 | 725 (153–1906) | 125/337 | 3.41 (2.36–4.99)* | 2.59 (1.76–3.85)* | 2.29 (1.49–3.55)* |
| faPWV, cm/s | |||||
| Quartile 1, 595–987.5 | 425 (45–1712) | 100/337 | 1 (reference) | 1 (reference) | 1 (reference) |
| Quartile 2, 988.5–1092 | 380 (42–1320) | 79/332 | 0.77 (0.55–1.08) | 0.73 (0.51–1.04) | 0.78 (0.53–1.14) |
| Quartile 3, 1092.5–1224.5 | 277 (32–1265) | 81/343 | 0.71 (0.50–1.00) | 0.57 (0.40–0.83)* | 0.56 (0.38–0.84)* |
| Quartile 4, 1225–5748.5 | 324 (28–1226) | 77/337 | 0.70 (0.49–0.99)* | 0.51 (0.35–0.73)* | 0.47 (0.30–0.71)* |
Model 1 was an unadjusted model. Model 2 was adjusted for covariates: age, sex, race, and field center. Model 3 was further adjusted for covariates other than Model 2: education level, body mass index, antihypertensive medication use, systolic blood pressure, diabetes, total and high‐density lipoprotein cholesterol level, estimated glomerular filtration rate, prior coronary heart disease, prior heart failure, prior stroke, current drinker, and current smoker. baPWV indicates brachial–ankle PWV; cfPWV, carotid–femoral PWV; faPWV, femoral–ankle PWV; haPWV, heart–ankle PWV; hcPWV, heart–carotid PWV; hfPWV, heart–femoral PWV; IQI, interquartile interval; and PWV, pulse wave velocity.
Indicate statistical significance.
When we modeled PWV measures continuously using restricted cubic splines for CAC and descending aorta calcification specifically, we observed continuous associations of cfPWV above ≈1000 cm/s with CAC and descending aorta calcification (Figure 2), with a sharper gradient for the latter. We observed largely similar patterns for hfPWV, haPWV, and baPWV (Figure S3). faPWV demonstrated an inverse association with both CAC and descending aorta calcification.
Figure 2. Adjusted odds ratio for high CAC and descending aorta calcification according to cfPWV modeled as restricted cubic spline.
The pink line and light pink shading indicate the odds ratio for high CAC (≥75th percentile) and 95% CI. The blue line and blue shading indicate the odds ratio for descending aorta calcification (≥75th percentile) and 95% CI. Histograms indicate the count of the study population. The reference value is the median of cfPWV in the lowest quartile (910 cm/s). The odds ratio was estimated using multivariable logistic regression models adjusted for age, sex, race, field center, education level, body mass index, antihypertensive medication use, systolic blood pressure, diabetes, total and high‐density lipoprotein cholesterol levels, estimated glomerular filtration rate, prior coronary heart disease, prior heart failure, prior stroke, current drinker, and current smoker (n=1351). The figure displays estimates for the range of 0.5 to 99.5 percentiles of cfPWV. CAC indicates coronary artery calcium; and cfPWV, carotid–femoral pulse wave velocity.
We confirmed similar results overall in sensitivity analyses. When we stratified by age, sex, race, and systolic blood pressure, the associations of cfPWV with CAC and descending aorta calcification were largely consistent (Figure 3), although we found statistically significant interactions by systolic blood pressure for descending aorta calcification (greater OR when systolic blood pressure <133 mm Hg versus ≥133 mm Hg, P‐for‐interaction=0.003). The results were consistent or even stronger than the primary analysis when we included ankle–brachial index and heart rate in our models (Figures S4 and S5). Exclusion of participants with severe obesity or prior revascularization, those with PWV measures of possible outliers, and those with aortic stenosis did not alter the results (Figures S6 through S8).
Figure 3. Association of cfPWV with CAC and descending aorta calcification among subgroups.
Adjusted odds ratio for high CAC score (A) and descending aorta calcification (B) of cfPWV 1‐SD increment was stratified according to baseline characteristics (age, sex, race, and systolic blood pressure). Cutoff value of age and systolic blood pressure was each median value. Odds ratio was estimated using multivariable logistic regression model adjusted for age, sex, race, field center, education level, body mass index, antihypertensive medication use, systolic blood pressure, diabetes, total and high‐density lipoprotein cholesterol levels, estimated glomerular filtration rate, prior coronary heart disease, prior heart failure, prior stroke, current drinker, and current smoker except for each stratification variable. Interaction was tested using likelihood ratio tests. CAC indicates coronary artery calcium; and cfPWV, carotid‐femoral pulse wave velocity.
Discrimination of High CAC and Descending Aorta Calcification
The C statistic of high CAC according to the base model (ie, covariates in Model 3) was 0.727 (95% CI, 0.697–0.758) (Table 4). When we added individual PWV measures separately, the largest improvement of discrimination was seen by cfPWV, although it was borderline significant (∆C statistic 0.008 [95% CI, −0.001 to 0.018]; P=0.07) (Table 4). When we added cfPWV with other PWV measures, we observed statistically significant improvement in the C statistic when we simultaneously added hfPWV, cfPWV, haPWV, and baPWV. Among these models, however, the lowest Akaike information criterion and Bayesian information criterion were seen when we added cfPWV to the base model.
Table 4.
C Statistics for High Coronary Artery Calcium Score (≥75th Percentile) by Adding PWV Measures to Prediction Model Based on Traditional Risk Factors
| Predictor | C statistic (95% CI) | AIC | BIC | |
|---|---|---|---|---|
| Base model | 0.727 (0.697–0.758) | 1392.036 | 1506.626 | |
| C statistic (95% CI) | C statistic difference from base model (95% CI) | |||
| +hcPWV | 0.728 (0.697 to 0.758) | 0.001 (−0.001 to 0.002) | 1393.559 | 1513.356 |
| +hfPWV | 0.731 (0.701 to 0.761) | 0.004 (−0.003 to 0.011) | 1386.404 | 1506.202 |
| +cfPWV | 0.736 (0.706 to 0.766) | 0.008 (−0.001 to 0.018) | 1379.592* | 1499.39* |
| +haPWV | 0.728 (0.697 to 0.758) | 0 (−0.003 to 0.003) | 1392.726 | 1512.524 |
| +baPWV | 0.728 (0.698 to 0.759) | 0.001 (−0.003 to 0.004) | 1391.985 | 1511.782 |
| +faPWV | 0.728 (0.697 to 0.758) | 0 (−0.001 to 0.002) | 1393.654 | 1513.452 |
| +hfPWV, cfPWV, haPWV, and baPWV | 0.738 (0.709 to 0.768)* | 0.011 (0.001 to 0.021)* | 1383.089 | 1518.512 |
| +hfPWV, cfPWV, haPWV, baPWV, and faPWV | 0.738 (0.708 to 0.768)* | 0.011 (0.001 to 0.021)* | 1384.884 | 1525.516 |
| +All PWVs | 0.738 (0.708 to 0.768)* | 0.011 (0.001 to 0.021)* | 1386.884 | 1532.725 |
Base model includes predictors: age, sex, race, field center, education level, body mass index, antihypertensive medication use, systolic blood pressure, diabetes, total and high‐density lipoprotein cholesterol level, estimated glomerular filtration rate, prior coronary heart disease, prior heart failure, prior stroke, current drinker, and current smoker. AIC indicates Akaike information criterion; baPWV, brachial–ankle PWV; BIC, Bayesian information criterion; cfPWV, carotid–femoral PWV; faPWV, femoral–ankle PWV; haPWV, heart–ankle PWV; hcPWV, heart–carotid PWV; hfPWV, heart–femoral PWV; and PWV, pulse wave velocity.
Statistical significance by the DeLong test or the lowest AIC and BIC.
For descending aorta calcification, the addition of cfPWV alone to the base model improved the C statistic by 0.021 (95% CI, 0.005–0.037) (Table S10). Further improvement was seen when we combined cfPWV with other PWVs. The lowest Akaike information criterion and Bayesian information criterion for descending aorta calcification were seen when we simultaneously added hfPWV, cfPWV, haPWV, and baPWV to the base model.
DISCUSSION
In this community‐based study of older populations with data on PWV measures and calcification of different vascular beds, we observed heterogeneous associations between specific segments of arterial stiffness and vascular beds of calcification. Among PWV measures, cfPWV was most strongly associated with vascular and valvular calcification, followed by hfPWV and haPWV. Interestingly, faPWV was inversely associated with vascular calcification. Among vascular beds tested, descending aorta calcification was most strongly associated with PWV measures. Some of these PWV measures modestly improved the discrimination of high levels of CAC and descending aorta calcification beyond traditional atherosclerotic risk factors.
There are several studies reporting associations between cfPWV and baPWV with CAC, 12 , 13 , 14 , 15 , 16 but our study is unique in several aspects. First, we reported that not only cfPWV, but also other PWV measures including hfPWV and haPWV, were associated with high CAC scores. Second, leveraging data on ECC, we found that most PWV measures were associated with different vascular beds tested, with the strongest association with descending aorta calcification. Finally, we rigorously evaluated whether the addition of cfPWV and other PWV measures can improve the discrimination of CAC or descending aortic calcification.
Our observation of cfPWV being most consistently associated with vascular calcification (including CAC) is in line with previous studies showing the prognostic value of cfPWV superior to other PWV measures 24 , 25 and further supports cfPWV as a gold standard measure of arterial stiffness. 2 Simultaneously, it is important to recognize that some other PWVs showed similar associations with CAC or descending aorta calcification (eg, hfPWV, haPWV, and baPWV). Importantly, some of these (eg, baPWV) are easier to measure than cfPWV, which requires tonometry sensors at the carotid and femoral and may be uncomfortable for some subjects. Thus, depending on research questions or clinical purpose, some of hfPWV, haPWV, and baPWV may be acceptable as an alternate of cfPWV. 26 Also, the repeatability of each PWV measure should be considered. 27 Furthermore, our discrimination analysis demonstrated additional values of different PWV measures, and thus whenever possible, the comprehensive assessment of multiple PWV measures may be beneficial. Although the amount is small, CAC measure requires radiation in an imaging center. Thus, if some individuals are reluctant to undergo CAC assessment, our study suggests that the PWV assessment in a clinic provides additional information on the probability of having higher CAC beyond traditional risk factors.
All PWV measures tested in the present study, except faPWV, reflect stiffness of the aorta to some degree. Thus, it is not surprising that among vascular beds tested, descending aorta calcification was most strongly associated with PWV measures in our study. Calcification of elastic fibers in the arterial wall contributes to arterial stiffness. 28 , 29 Although several studies report the association of thoracic aorta calcification (ascending aorta, aortic arch, and descending aorta) with adverse events, 7 , 10 , 11 , 30 , 31 , 32 compared with CAC score, clinical usefulness of descending aorta calcification is less established. If prognostic value of descending aorta calcification beyond CAC is established in the future, cfPWV alone (or in combination with a few other PWV measures) would be helpful to identify individuals who are likely to have descending aorta calcification.
In our study, faPWV showed an inverse association with descending aorta calcification. We have shown a similar inverse association of faPWV with cardiac biomarkers, kidney function, and cardiovascular risk factors previously. 19 , 33 , 34 We are not sure why higher faPWV is consistently associated with better clinical phenotypes. These counterintuitive observations may reflect different histology or structure between aorta and peripheral arteries. For example, shorter wave transit time behind higher PWV may have different pathophysiological meanings between an elastic artery like the aorta versus a muscular artery in the periphery. Nonetheless, because faPWV has been much less studied compared with other PWV measures, dedicated investigation on faPWV (eg, whether higher faPWV represent greater leg artery calcification) may help us better understand pathophysiology across different vascular beds.
Our study has several limitations. First, our study included older Black and White participants, and thus, the extrapolation of our results to other age ranges or different racial and ethnic groups needs careful discussion. Second, in some participants, the elapsed time between PWV measurement and CT scan was >1 year. Third, because the images of vascular beds tested were acquired from the level of the pulmonary artery bifurcation to the apex of the heart, a few other potentially relevant vascular beds (eg, aortic arch, abdominal aorta, leg artery) were not assessed in the present study. Fourth, we are not sure whether our findings are transportable to PWV measures evaluated by different measurement system. Finally, although we did our best to adjust for potential confounders, we cannot deny the possibility of residual confounding due to unmeasured or unknown confounders.
In conclusion, several PWV measures, particularly cfPWV, were independently associated with CAC and calcification of other vascular beds, such as descending aorta calcification, in older adults. Among different vascular beds, descending aorta calcification was most strongly associated with PWV measures tested. Our study further supports unique pathophysiological process across different vascular beds and suggests that some PWV measures, especially cfPWV, could help identify individuals with CAC or descending aorta calcification.
Sources of Funding
The ARIC study is performed as a collaborative study supported by National Heart, Lung, and Blood Institute contracts (75N92022D00001, 75N92022D00002, 75N92022D00003, 75N92022D00004, 75N92022D00005, and 75N92022D00005). The funders of this study had no role in study design; collection, analysis, and interpretation of data; writing the report; and the decision to submit the report for publication. Arterial stiffness measures were funded by R01AG053938. The present study was supported by R01HL136592 (Co‐Principal Investigators: K.M. and M.J.B.).
Disclosures
Dr Matsushita received honorarium from Fukuda Denshi outside of the submitted work.
Supporting information
Data S1
Tables S1–S10
Figures S1–S8
Reference 35
Acknowledgments
The authors thank the staff and participants of the ARIC study for their important contributions.
Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.123.031778
This article was sent to Hani Jneid, MD, Associate Editor, for review by expert referees, editorial decision, and final disposition.
This study was presented in part at the European Society of Cardiology Congress, August 28, 2022, in Barcelona, Spain, and published in abstract form in the European Heart Journal, Vol 43: Issue Supplement 2. doi: 10.1093/eurheartj/ehac544.186.
For Sources of Funding and Disclosures, see page 11.
References
- 1. Boutouyrie P, Chowienczyk P, Humphrey JD, Mitchell GF. Arterial stiffness and cardiovascular risk in hypertension. Circ Res. 2021;128:864–886. doi: 10.1161/circresaha.121.318061 [DOI] [PubMed] [Google Scholar]
- 2. Townsend RR, Wilkinson IB, Schiffrin EL, Avolio AP, Chirinos JA, Cockcroft JR, Heffernan KS, Lakatta EG, McEniery CM, Mitchell GF, et al. Recommendations for improving and standardizing vascular research on arterial stiffness: a scientific statement from the American Heart Association. Hypertension. 2015;66:698–722. doi: 10.1161/hyp.0000000000000033 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Salvi P, Scalise F, Rovina M, Moretti F, Salvi L, Grillo A, Gao L, Baldi C, Faini A, Furlanis G, et al. Noninvasive estimation of aortic stiffness through different approaches. Hypertension. 2019;74:117–129. doi: 10.1161/hypertensionaha.119.12853 [DOI] [PubMed] [Google Scholar]
- 4. Vlachopoulos C, Aznaouridis K, Stefanadis C. Prediction of cardiovascular events and all‐cause mortality with arterial stiffness: a systematic review and meta‐analysis. J Am Coll Cardiol. 2010;55:1318–1327. doi: 10.1016/j.jacc.2009.10.061 [DOI] [PubMed] [Google Scholar]
- 5. Greenland P, Blaha MJ, Budoff MJ, Erbel R, Watson KE. Coronary calcium score and cardiovascular risk. J Am Coll Cardiol. 2018;72:434–447. doi: 10.1016/j.jacc.2018.05.027 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Grundy SM, Stone NJ, Bailey AL, Beam C, Birtcher KK, Blumenthal RS, Braun LT, de Ferranti S, Faiella‐Tommasino J, Forman DE, et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA guideline on the management of blood cholesterol: a report of the American College of Cardiology/American Heart Association task force on clinical practice guidelines. Circulation. 2019;139:e1082–e1143. doi: 10.1161/cir.0000000000000625 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Budoff MJ, Nasir K, Katz R, Takasu J, Carr JJ, Wong ND, Allison M, Lima JA, Detrano R, Blumenthal RS, et al. Thoracic aortic calcification and coronary heart disease events: the multi‐ethnic study of atherosclerosis (MESA). Atherosclerosis. 2011;215:196–202. doi: 10.1016/j.atherosclerosis.2010.11.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Churchill TW, Rasania SP, Rafeek H, Mulvey CK, Terembula K, Ferrari V, Jha S, Lilly SM, Eraso LH, Reilly MP, et al. Ascending and descending thoracic aorta calcification in type 2 diabetes mellitus. J Cardiovasc Comput Tomogr. 2015;9:373–381. doi: 10.1016/j.jcct.2015.04.008 [DOI] [PubMed] [Google Scholar]
- 9. Hoffmann U, Massaro JM, D'Agostino RB Sr, Kathiresan S, Fox CS, O'Donnell CJ. Cardiovascular event prediction and risk reclassification by coronary, aortic, and valvular calcification in the Framingham heart study. J Am Heart Assoc. 2016;5:5. doi: 10.1161/jaha.115.003144 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Cho IJ, Chang HJ, Cho I, Heo R, Lee SE, Shim CY, Hong GR, Chung N. Association of thoracic aorta calcium score with exercise blood pressure response and clinical outcomes in elderly individuals: differential impact of aorta calcification compared with coronary artery calcification. J Am Heart Assoc. 2016;5:5. doi: 10.1161/jaha.115.003131 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Dudink E, Peeters F, Altintas S, Heckman LIB, Haest RJ, Kragten H, Kietselaer B, Wildberger J, Luermans J, Weijs B, et al. Agatston score of the descending aorta is independently associated with coronary events in a low‐risk population. Open Heart. 2018;5:e000893. doi: 10.1136/openhrt-2018-000893 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Kullo IJ, Bielak LF, Turner ST, Sheedy PF II, Peyser PA. Aortic pulse wave velocity is associated with the presence and quantity of coronary artery calcium: a community‐based study. Hypertension. 2006;47:174–179. doi: 10.1161/01.Hyp.0000199605.35173.14 [DOI] [PubMed] [Google Scholar]
- 13. van Popele NM, Mattace‐Raso FU, Vliegenthart R, Grobbee DE, Asmar R, van der Kuip DA, Hofman A, de Feijter PJ, Oudkerk M, Witteman JC. Aortic stiffness is associated with atherosclerosis of the coronary arteries in older adults: the Rotterdam study. J Hypertens. 2006;24:2371–2376. doi: 10.1097/01.hjh.0000251896.62873.c4 [DOI] [PubMed] [Google Scholar]
- 14. Raggi P, Bellasi A, Ferramosca E, Islam T, Muntner P, Block GA. Association of pulse wave velocity with vascular and valvular calcification in hemodialysis patients. Kidney Int. 2007;71:802–807. doi: 10.1038/sj.ki.5002164 [DOI] [PubMed] [Google Scholar]
- 15. Cainzos‐Achirica M, Rampal S, Chang Y, Ryu S, Zhang Y, Zhao D, Cho J, Choi Y, Pastor‐Barriuso R, Lim SY, et al. Brachial‐ankle pulse wave velocity is associated with coronary calcium in young and middle‐aged asymptomatic adults: the Kangbuk Samsung health study. Atherosclerosis. 2015;241:350–356. doi: 10.1016/j.atherosclerosis.2015.05.031 [DOI] [PubMed] [Google Scholar]
- 16. Lee JY, Ryu S, Lee SH, Kim BJ, Kim BS, Kang JH, Cheong ES, Kim JY, Park JB, Sung KC. Association between brachial‐ankle pulse wave velocity and progression of coronary artery calcium: a prospective cohort study. Cardiovasc Diabetol. 2015;14:147. doi: 10.1186/s12933-015-0311-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Wright JD, Folsom AR, Coresh J, Sharrett AR, Couper D, Wagenknecht LE, Mosley TH Jr, Ballantyne CM, Boerwinkle EA, Rosamond WD, et al. The ARIC (atherosclerosis risk in communities) study: JACC focus seminar 3/8. J Am Coll Cardiol. 2021;77:2939–2959. doi: 10.1016/j.jacc.2021.04.035 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Cortez‐Cooper MY, Supak JA, Tanaka H. A new device for automatic measurements of arterial stiffness and ankle‐brachial index. Am J Cardiol. 2003;91:1519–1522. doi: 10.1016/s0002-9149(03)00416-8 [DOI] [PubMed] [Google Scholar]
- 19. Kim ED, Tanaka H, Ballew SH, Sang Y, Heiss G, Coresh J, Matsushita K. Associations between kidney disease measures and regional pulse wave velocity in a large community‐based cohort: the atherosclerosis risk in communities (ARIC) study. Am J Kidney Dis. 2018;72:682–690. doi: 10.1053/j.ajkd.2018.04.018 [DOI] [PubMed] [Google Scholar]
- 20. Sugawara J, Hayashi K, Yokoi T, Cortez‐Cooper MY, DeVan AE, Anton MA, Tanaka H. Brachial‐ankle pulse wave velocity: an index of central arterial stiffness? J Hum Hypertens. 2005;19:401–406. doi: 10.1038/sj.jhh.1001838 [DOI] [PubMed] [Google Scholar]
- 21. Obisesan OH, Kou M, Wang FM, Boakye E, Honda Y, Uddin SMI, Dzaye O, Osei AD, Orimoloye OA, Howard‐Claudio CM, et al. Lipoprotein(a) and subclinical vascular and valvular calcification on cardiac computed tomography: the atherosclerosis risk in communities study. J Am Heart Assoc. 2022;11:e024870. doi: 10.1161/jaha.121.024870 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M Jr, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol. 1990;15:827–832. doi: 10.1016/0735-1097(90)90282-t [DOI] [PubMed] [Google Scholar]
- 23. Inker LA, Eneanya ND, Coresh J, Tighiouart H, Wang D, Sang Y, Crews DC, Doria A, Estrella MM, Froissart M, et al. New creatinine‐ and cystatin C‐based equations to estimate GFR without race. N Engl J Med. 2021;385:1737–1749. doi: 10.1056/NEJMoa2102953 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. 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: 10.1016/j.jacc.2013.09.063 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Kim ED, Ballew SH, Tanaka H, Heiss G, Coresh J, Matsushita K. Short‐term prognostic impact of arterial stiffness in older adults without prevalent cardiovascular disease. Hypertension. 2019;74:1373–1382. doi: 10.1161/hypertensionaha.119.13496 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Stoner L, Meyer ML, Kucharska‐Newton A, Stone K, Zieff G, Dave G, Fryer S, Credeur D, Faulkner J, Matsushita K, et al. Associations between carotid‐femoral and heart‐femoral pulse wave velocity in older adults: the atherosclerosis risk in communities study. J Hypertens. 2020;38:1786–1793. doi: 10.1097/hjh.0000000000002449 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Meyer ML, Tanaka H, Palta P, Patel MD, Camplain R, Couper D, Cheng S, Al Qunaibet A, Poon AK, Heiss G. Repeatability of central and peripheral pulse wave velocity measures: the atherosclerosis risk in communities (ARIC) study. Am J Hypertens. 2016;29:470–475. doi: 10.1093/ajh/hpv127 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Mackey RH, Venkitachalam L, Sutton‐Tyrrell K. Calcifications, arterial stiffness and atherosclerosis. Adv Cardiol. 2007;44:234–244. doi: 10.1159/000096744 [DOI] [PubMed] [Google Scholar]
- 29. Blaha MJ, Budoff MJ, Rivera JJ, Katz R, O'Leary DH, Polak JF, Takasu J, Blumenthal RS, Nasir K. Relationship of carotid distensibility and thoracic aorta calcification: multi‐ethnic study of atherosclerosis. Hypertension. 2009;54:1408–1415. doi: 10.1161/hypertensionaha.109.138396 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Santos RD, Rumberger JA, Budoff MJ, Shaw LJ, Orakzai SH, Berman D, Raggi P, Blumenthal RS, Nasir K. Thoracic aorta calcification detected by electron beam tomography predicts all‐cause mortality. Atherosclerosis. 2010;209:131–135. doi: 10.1016/j.atherosclerosis.2009.08.025 [DOI] [PubMed] [Google Scholar]
- 31. Lessmann N, de Jong PA, Celeng C, Takx RAP, Viergever MA, van Ginneken B, Išgum I. Sex differences in coronary artery and thoracic aorta calcification and their association with cardiovascular mortality in heavy smokers. JACC Cardiovasc Imaging. 2019;12:1808–1817. doi: 10.1016/j.jcmg.2018.10.026 [DOI] [PubMed] [Google Scholar]
- 32. Castagna F, Miles J, Arce J, Leiderman E, Neshiwat P, Ippolito P, Friedmann P, Schenone A, Zhang L, Rodriguez CJ, et al. Visual coronary and aortic calcium scoring on chest computed tomography predict mortality in patients with low‐density lipoprotein‐cholesterol ≥190 mg/dL. Circ Cardiovasc Imaging. 2022;15:e014135. doi: 10.1161/circimaging.122.014135 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Liu S, Kim ED, Wu A, Meyer ML, Cheng S, Hoogeveen RC, Ballantyne CM, Tanaka H, Heiss G, Selvin E, et al. Central and peripheral pulse wave velocity and subclinical myocardial stress and damage in older adults. PLoS One. 2019;14:e0212892. doi: 10.1371/journal.pone.0212892 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Meyer ML, Tanaka H, Palta P, Cheng S, Gouskova N, Aguilar D, Heiss G. Correlates of segmental pulse wave velocity in older adults: the atherosclerosis risk in communities (ARIC) study. Am J Hypertens. 2016;29:114–122. doi: 10.1093/ajh/hpv079 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Yamashina A, Tomiyama H, Takeda K, Tsuda H, Arai T, Hirose K, Koji Y, Hori S, Yamamoto Y. Validity, reproducibility, and clinical significance of noninvasive brachial‐ankle pulse wave velocity measurement. Hypertens Res. 2002;25:359–364. doi: 10.1291/hypres.25.359 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1
Tables S1–S10
Figures S1–S8
Reference 35