Abstract
Background
The association between measures of arterial compliance and peripheral arterial disease (PAD) is unclear. Early changes in arterial wall compliance could be a useful marker of patients at high risk for developing lower extremity atherosclerosis.
Methods
We used linear and logistic regression models on baseline data from 2803 female and 2558 male participants in the Multi-Ethnic Study of Atherosclerosis (MESA) to study associations between tonometry-derived baseline measures of arterial compliance (large artery compliance [C1] and small artery compliance [C2]) and the baseline ankle-brachial index (ABI), as well as change in the ABI over approximately 3 years of follow up.
Results
In cross-sectional analyses, lower C1 and C2 values, indicating poorer arterial compliance, were associated with lower ABI. There were significant linear trends across strata of ABI, especially in C2 which ranged from 3.7ml/mmHg × 100 (95% confidence interval (CI) 3.3 to 4.2) in women with an ABI < 0.90 to 4.2ml/mmHg × 100 (95% CI 4.1 to 4.3 p<0.001) in women with ABI 1.10 - <1.40. Similar significant trends (p<0.001) were seen in men. In prospective analyses, those with the lowest tertile of C2 values at baseline had a greater multivariable-adjusted odds for decline in ABI of ≥ 0.15 over 3 years compared to those with the highest C2 values at baseline (OR 1.80 95% CI 1.23–2.64).
Conclusions
We observed that less compliant arteries were significantly associated with low ABI in cross-sectional analysis and with greater decline in ABI over time.
Keywords: Ankle-Brachial Index, Arterial Compliance, Peripheral Arterial Disease
Introduction
The ankle-brachial index (ABI) is a simple and non-invasive method for diagnosing and assessing severity of lower extremity peripheral arterial disease (PAD). Persons without lower extremity atherosclerosis typically have an ABI > 1.00, because of increasing blood pressures in distal vasculature due to increased arterial taper. 1 An ABI < 0.90 shows good sensitivity and excellent specificity for a lower extremity arterial stenosis >50% on digital subtraction angiography.2 However, ABI <0.90 is a relatively late manifestation of atherosclerosis. Prior to the development of an occlusive atheroma, other functional changes in arterial wall function and structure may occur.3, 4 Identifying early changes in arterial wall function and structure could be a useful indicator of sub-clinical vascular disease in patients at high risk for developing lower extremity atherosclerosis. Early identification of patients with lower extremity atherosclerosis is important as people with PAD are at increased risk for cardiovascular events, functional decline, and they represent a potential target population for early preventive intervention.5–7
Peripheral arterial pulse wave analysis, derived from radial artery applanation tonometry, is a non-invasive technique for measuring arterial compliance. Systemic measures of large vessel compliance, C1, and small vessel compliance, C2, can be derived from this method. 8–10 Lower C2 values, indicating poorer small vessel compliance, have been observed in subjects with older age, type II diabetes mellitus, hypertension, tobacco use, and in post menopausal women with coronary heart disease (CHD). Recently, prospective data demonstrate independent associations between lower C2 and incident hypertension, CHD events and stroke.11, 12 No such associations have been observed for C1.13–16 Since atherosclerosis is a systemic disease, alterations in radial artery pulse wave characteristics may be seen in patients with low ABI even though the ABI is largely dependent upon lower-extremity blood pressure. Though limited, current data suggest there may be some relationship between tonometry-derived measures of arterial compliance and the existence of PAD. 17 Weak cross-sectional associations have been observed between C2 and ABI in diabetic patients, but data from large community-based samples are lacking.13, 17 To our knowledge, no prior studies have investigated longitudinal relationships between baseline arterial compliance measures and change in ABI.
We studied associations between C1 and C2 with presence and severity of PAD among participants in the Multi-Ethnic Study of Atherosclerosis (MESA) cohort. We also assessed associations of these compliance measures at baseline with changes in the ABI over time. If lower arterial compliance is associated with lower ABI values and greater declines in the ABI over time, applanation tonometry derived arterial compliance measures could be a non-invasive technique to identify early subclinical atherosclerosis in the lower extremities. Such associations could provide mechanistic insight into factors mediating the development and progression of PAD. We hypothesized that lower C1 and C2 values indicating lower arterial compliance, are associated with lower ABI values at baseline. We further hypothesized that lower C1 and C2 values at baseline would be associated with greater decline in the ABI at 3 year follow up.
Methods
Participants
The MESA cohort has been described previously 18 In brief, between July 2000 and August 2002, 6814 men and women who identified themselves as White, African American, Hispanic, or Chinese, were 45 to 84 years old, and were free of clinically apparent cardiovascular disease (CVD) were recruited and participated in the baseline examination. Individuals with a history of physician-diagnosed heart attack, angina, heart failure, stroke, or transient ischemic attack (TIA), or having undergone an invasive procedure for cardiovascular disease were excluded from participation. Participants who underwent lower extremity revascularization during follow up and individuals with ABI > 1.40 at baseline were excluded from this analysis. The participants enrolled at baseline were eligible to participate in three subsequent study examinations occurring approximately every 18 to 24 months and referred to as clinic visits 2, 3, and 4.
Participants were recruited from the following six communities: Baltimore City and Baltimore County, MD; Chicago, IL; Forsyth County, NC; Los Angeles County, CA; Northern Manhattan and the Bronx, NY; and St. Paul, MN. The institutional review boards at all participating centers approved the study and all participants gave informed written consent.
Data Collection
At the baseline examination standardized questionnaires were used to obtain demographic information and medication usage for high blood pressure, high cholesterol, or diabetes. Cigarette smoking was calculated in pack-years and also defined as current, former, or never. Height and weight were measured with participants wearing light clothing and no shoes. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters. Resting blood pressure was measured three times in seated participants with a Dinamap model Pro 100 automated oscillometric sphygmomanometer (Critikon, Tampa, Fla). The average of the last two measurements was used in the analysis. Hypertension was defined as the systolic blood pressure ≥ 140mm Hg, diastolic blood pressure ≥ 90mm Hg, or current use of an antihypertensive medication.
Laboratory
At the baseline examination and at visit 3, blood was collected after a 12-hour fast and stored at −70 degrees centigrade. Total and high-density lipoprotein (HDL) cholesterol, triglycerides, glucose and creatinine were measured and as previously reported low-density lipoprotein (LDL) cholesterol was calculated by the Friedwald equation.19 Diabetes was defined as fasting glucose ≥ 126mg/dL or use of glucose -lowering medication. 20 The estimated glomerular filtration rate (eGFR) was calculated using the modified of diet in renal disease (MDRD) study formula. 21
Ankle-brachial index protocol
At the baseline examination and Exam 3, systolic blood pressure measurements for calculation of the ABI were obtained using a hand-held Doppler instrument with a 5-mHz probe (Nicolet Vascular, Golden, Colo). In brief, systolic blood pressures were measured in the bilateral brachial, dorsalis pedis, and posterior tibial arteries. Brachial artery pressures were averaged to obtain the ABI denominator. When the two brachial artery pressures differed by 10mmHg or more the higher brachial artery pressure was used as the denominator. For each lower extremity, the ABI numerator used was the highest pressure (dorsalis pedis or posterior tibial) from that leg. The lower of the right and left ABI values was used as the index ABI for that subject unless one leg yielded an ABI >1.40 in which case the higher ABI was used and the participant was excluded from these analyses. In the longitudinal analyses, the ABIs from the same leg were compared from baseline to Exam 3. We defined decline in ABI as a decrease in ABI > 0.15; increase in ABI was defined as an increase in ABI > 0.15, and stable ABI was defined as change ≤0.15. We chose the cut-point of 0.15 because this magnitude of change in ABI has been used to define significant progression of PAD in previous studies. Furthermore, independent associations with increased PAD symptoms and all cause mortality and ABI decline >0.15 have been demonstrated as well,22–24
Determination of C1 and C2
At the baseline examination, non-invasive measures of arterial stiffness for calculating C1 and C2 were obtained with the Hypertension Diagnostics, Inc. Pulse Wave CR-2000 device (Eagan, Minnesota, USA). The Pulse Wave CR-2000 obtains blood pressure measurements and arterial pulse waveforms with an oscillometric blood pressure module, and an arterial pulse pressure sensor. The blood pressure measurements are taken from the upper left arm and the pulse pressure sensor is positioned on the right wrist. Participants were supine and their right wrist was placed in the wrist stabilizer. The tonometry sensor was placed over the area of maximum radial artery pulsation. The HDI Pulse Wave CR-2000 device makes automatic adjustments to ensure a stable waveform. A modified Windkessel model, which is analogous to an electrical circuit, is used to generate systemic arterial compliance measures. The model calculates a capacitive compliance measure, C1, to represent large arterial vessel compliance and an oscillatory or reflective element, C2, to represent small arterial vessel compliance. Several coefficients for the Windkessel formula were derived from the diastolic portion of the pressure wave form. The technique for coefficient derivation is proprietary and was unavailable to the research team. These coefficients are then multiplied by an estimated systemic vascular resistance to obtain the capacitive compliance of the proximal aorta (C1) and oscillatory compliance of small distal arteries (C2) values which represent large and small vessel compliance respectively.25 Systemic vascular resistance is estimated by dividing the mean arterial pressure by an estimated cardiac output. Cardiac output was estimated from ejection time, body surface area, and heart rate.12
Statistical Analyses
Of the 5496 MESA participants, 135 were excluded from the cross-sectional analysis, due to missing data for either baseline ABI, C1, C2, clinical covariates, or follow up ABI measurements. A total of 5361 had ABI measurements at both baseline and Exam 3, baseline pulse wave measures, and all covariates included in our final model.
ABI was stratified according to previously published definitions: ABI < 0.90 (definite PAD); 0.90 < ABI < 1.00 (borderline ABI); 1.00 < ABI < 1.10 (low-normal ABI); 1.10 < ABI < 1.40 (normal ABI). 26 Arterial compliance measures were analyzed as continuous variables and also were categorized into low, middle, and high according to tertiles of C1 and C2 distribution within the entire cohort. All analyses by baseline ABI groups were performed separately for women and men. Baseline characteristics of study participants were presented as mean (SD) for continuous variables, or percentage for categorical variables, and linear trends across the 4 ABI groups was tested with ABI as a continuous variable using linear regression (for continuous variables) or Cochran-Armitage trend test (for categorical variables). In the cross-sectional analyses, general linear models were used to compute ABI group mean compliance measures, adjusted for baseline age, race, cigarette smoking, diabetes, hypertension, eGFR, and total/HDL cholesterol ratio. Pairwise comparison were made between the normal ABI group (referent) and the PAD, borderline, or low-normal ABI groups, and testing for linear trend was performed with continuous ABI using the general linear model.
For the longitudinal analysis, change in ABI over 3 year follow-up was categorized into 3 groups: ABI decline ≥ 0.15; stable ABI (change ≤ +/− 0.15), and ABI increase > 0.15.27 C1 and C2 values were examined continuously and as tertiles (low, middle, and high). Separate linear and logistic regression models were used to examine the odds ratios (OR) and 95% confidence intervals (CI) of ABI progression (ABI decrease or increase ≥ 0.15), compared to people maintaining a stable ABI (ABI change < 0.15). In the multivariate models, we first adjusted for age, gender, race, cigarette smoking, hypertension, diabetes, eGFR, total/HDL cholesterol ratio, heart rate and baseline ABI. We further adjusted for height and weight. We then stratified by baseline ABI to ensure there was no residual confounding due to this variable. With stratification there was no significant change in our results, thus we pooled all baseline ABI groups and adjusted for baseline ABI in our multivariable models. Multivariable adjusted logistic regression was used to test differences between the arterial compliance high tertile (referent) and the low or middle arterial compliance tertile. Linear trend across the arterial compliance tertiles was tested by entering natural log-transformed C1 or C2 variables in logistic regression. All analyses were performed using SAS statistical software Version 9.2 (SAS institute, Cary, NC). A p value of < 0.05 was considered statistically significant.
Results
Baseline Characteristics
Baseline characteristics for female and male participants are shown in Table 1. As expected, participants with PAD had a higher mean age and higher prevalence of systolic hypertension, tobacco use, and diabetes compared to participants with a normal ABI.
Table 1.
Baseline Characteristics of Women and Men According to Ankle-Brachial Index (ABI), MESA (2000–2002)
| Women (N=2803) | Men (N=2558) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| ABI | ABI | |||||||||
| Definite PAD (<0.90) |
Borderline ABI (0.90-<1.00) |
Low-normal ABI (1.00-<1.10) |
Normal ABI (1.10-<1.40) |
P trend | Definite PAD (<0.90) |
Borderline ABI (0.90-<1.00) |
Low-normal ABI (1.00-<1.10) |
Normal ABI (1.10-<1.40) |
P trend | |
| N | 81 | 270 | 1047 | 1405 | 73 | 105 | 552 | 1828 | ||
| Age (years) | 70.2 (9.9) | 64.2 (10.2) | 62.2 (10.0) | 59.6 (9.6) | <0.001 | 69.5 (9.1) | 66.2 (10.1) | 62.6 (10.2) | 60.7 (9.8) | <0.001 |
| Body mass index (kg/m2) | 28.4 (5.7) | 28.3 (5.9) | 28.1 (5.8) | 29.0 (6.3) | <0.001 | 27.2 (3.9) | 28.1 (4.8) | 27.2 (4.3) | 28.0 (4.3) | <0.001 |
| Smoking (%) | ||||||||||
| Current | 14.8 | 16.7 | 10.4 | 10.2 | 0.011 | 27.4 | 19.0 | 18.1 | 11.3 | <0.001 |
| Never | 51.8 | 53.3 | 60.1 | 61.7 | 0.006 | 20.5 | 26.7 | 36.6 | 44.6 | <0.001 |
| Blood pressure (mmHg) | ||||||||||
| Systolic | 141.5 (28.7) | 133.2 (23.0) | 128.8 (23.3) | 120.7 (20.2) | <0.001 | 136.6 (23.2) | 132.5 (19.7) | 127.0 (19.4) | 124.1 (18.0) | <0.001 |
| Diastolic | 70.5 (12.1) | 70.3 (11.1) | 69.6 (10.3) | 68.0 (9.6) | <0.001 | 74.2 (10.3) | 75.4 (9.1) | 75.4 (9.5) | 75.1 (9.1) | 0.199 |
| Medication Use (%) | ||||||||||
| Blood Pressure | 54.3 | 45.2 | 39.2 | 31.2 | <0.001 | 58.9 | 45.7 | 36.6 | 32.5 | <0.001 |
| Statin | 30.9 | 19.3 | 14.5 | 12.7 | <0.001 | 20.5 | 20.9 | 14.3 | 13.8 | 0.030 |
| Diabetes (%) | 30.9 | 8.1 | 9.9 | 9.7 | 0.005 | 19.2 | 14.3 | 12.3 | 11.9 | 0.106 |
| Cholesterol (mg/dL) | ||||||||||
| Total | 203.0 (32.8) | 197.3 (35.5) | 200.1 (34.2) | 199.1 (34.5) | 0.274 | 187.6 (35.4) | 191.1 (35.0) | 189.2 (38.0) | 187.6 (33.4) | 0.082 |
| HDL | 53.0 (14.1) | 56.5 (15.2) | 57.0 (15.9) | 56.4 (14.7) | 0.747 | 44.4 (10.9) | 46.3 (12.8) | 45.3 (12.0) | 44.9 (11.4) | 0.605 |
| Total/HDL ratio | 4.1 (1.3) | 3.7 (1.1) | 3.7 (1.1) | 3.7 (1.1) | 0.120 | 4.4 (1.2) | 4.4 (1.2) | 4.4 (1.5) | 4.4 (1.2) | 0.265 |
| GFR | 73.0 (20.6) | 79.1 (17.0) | 79.9 (17.0) | 81.0 (17.1) | 0.001 | 82.3 (18.6) | 81.4 (19.7) | 83.7 (18.1) | 82.0 (16.4) | 0.876 |
Data are expressed as mean (SD) unless otherwiseindicated.
Cross sectional comparisons of C1 and C2 between different ABI categories at baseline
For both men and women, significant linear trends were observed at baseline across all ABI strata for C1 and C2. In fully adjusted analyses, C1 and C2 values, indicating poorer large and small vessel compliance, respectively, were lower in participants with lower ABI. Among women, C1 and C2 were significantly lower in the borderline ABI category as compared to the normal ABI category (Table 2).
Table 2.
Baseline Measures of Arterial Compliance by Ankle-brachial Index (ABI) Category*
| Women (N=2803) | Men (N=2558) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| ABI | P trend | ABI | P trend | |||||||
| Definite PAD (<0.90) N=81 | Borderline ABI (0.90-<1.00) N=270 | Low-normal ABI (1.00-<1.10) N=1047 | Normal ABI (1.10-<1.40) N=1405 | Definite PAD (<0.90) N=73 | Borderline ABI (0.90-<1.00) N=105 | Low-normal ABI (1.00-<1.10) N=552 | Normal ABI (1.10-<1.40) N=1828 | |||
| Large artery compliance (C1) (ml/mmHg ×10) | 11.8 (10.8, 12.8) | 11.2 ‡ (10.6, 11.7) | 11.9 (11.6, 12.1) | 12.1 (11.8, 12.3) | 0.008 | 14.7 (13.5, 15.9) | 15.1 (14.1, 16.1) | 14.8‡ (14.4, 15.2) | 15.5 (15.3, 15.7) | 0.001 |
| Small artery compliance (C2) (ml/mmHg ×100) | 3.7 (3.3, 4.2) | 3.4† (3.2, 3.6) | 3.6† (3.4, 3.7) | 4.2 (4.1, 4.3) | <0.001 | 4.3† (3.7, 4.9) | 4.6† (4.1, 5.1) | 4.9† (4.6, 5.1) | 5.6 (5.5, 5.7) | <0.001 |
Data are expressed as mean (95% CI) unless otherwise indicated. Analyses adjusted for age, race, cigarette smoking, diabetes, hypertension, GFR, and Total/ HDL cholesterol ratio.
p < 0.001,
p < 0.01, comparing to the reference group (ABI 1.10-<1.40).
Among men, C1 was not significantly lower across all low ABI strata in pairwise comparisons. C2, however, was significantly lower across all lower ABI strata when compared to the normal ABI referent.
Prospective associations between baseline PWA measures of arterial compliance and change in ABI
Among 5361 participants with three year follow-up data, 256 had a decline in ABI > 0.15. Because relatively few women had this magnitude of ABI decline, we pooled all participants and adjusted for sex. Baseline characteristics of participants in different groups of ABI change are presented in Table 3. When compared with the stable ABI group, participants with a decline in ABI > 0.15 were older, had a higher baseline ABI, and included a higher prevalence of cigarette smokers. Baseline characteristics associated with tertiles of C2 are shown in the supplementary table. In fully adjusted models, (See Table 4), the lowest C2 tertile at baseline was associated with an increased risk for decline in ABI > 0.15, compared to the stable ABI group (OR 1.80 95% CI 1.23–2.64). No significant associations between baseline C1 tertiles and ABI decline were identified.
Table 3.
Associations of Baseline Characteristics with 3-year ABI Change *
| ABI change at year 3 follow-up, relative to baseline | |||
|---|---|---|---|
| Decline ≥0.15 n=256 | Change <0.15 (Referent) n=4537 | Increase ≥ 0.15 n=568 | |
| Age (years) | 63.3 (10.4) ‡ | 61.5 (10.0) | 60.7 (10.1) |
| Male, N (%) | 125 (48.8) | 2167 (47.8) | 266 (46.8) |
| Ankle-brachial index | 1.17 (0.12) † | 1.12 (0.10) | 1.05 (0.12) † |
| Race, N (%) | |||
| White | 99 (38.7) | 1780 (39.2) | 239 (42.) |
| Chinese | 23 (9.0) | 571 (12.6) | 65 (11.4) |
| African American | 77 (30.1) | 1173 (25.8) | 167 (29.4) |
| Hispanic | 57 (22.3) | 1013 (22.3) | 97 (17.1) ‡ |
| Body mass index (kg/m2) | 28.7 (6.2) | 28.1 (5.2) | 29.1 (5.8) † |
| Smoking, N (%) | |||
| Current | 47 (18.4) ‡ | 551 (12.1) | 58 (10.2) |
| Never | 116 (45.3) | 2321 (51.2) | 306 (53.9) |
| Blood pressure (mmHg) | |||
| Systolic | 126.4 (22.8) | 125.2 (20.6) | 126.8 (21.3) |
| Diastolic | 71.0 (10.3) | 71.9 (10.1) | 72.4 (11.1) |
| Blood Pressure medication use, N (%) | 102 (39.8) | 1590 (35.0) | 211 (37.1) |
| Hypertension, N (%) | 118 (46.1) | 1925 (42.4) | 255 (44.9) |
| Diabetes, N (%) | 38 (14.8) | 504 (11.1) | 62 (10.9) |
| Total/HDL cholesterol ratio | 4.1 (1.3) | 4.1 (1.2) | 4.0 (1.2) |
| Glomerular filtration rate | 79.7 (18.6) | 81.2 (17.1) | 81.8 (16.7) |
| Pulse wave analysis | |||
| Large vessel compliance (C1) (ml/mmHg × 10) | 13.1 (5.9) | 13.5 (5.5) | 13.7 (5.9) |
| Small vessel compliance (C2) (ml/mmHg × 100) | 4.2 (2.6) § | 4.6 (2.8) | 4.7 (3.1) |
Data are expressed as mean (SD) unless otherwise indicated.
p < 0.001,
p < 0.01,
p < 0.05
Table 4.
Odds Ratios for ABI Decline ≥ 0.15 over 3 years according to tertiles of baseline arterial compliance among all participants *
| ABI decline >= 0.15 at year 3 (all participants) | ||||
|---|---|---|---|---|
| Large vessel compliance (C1) | Small vessel compliance (C2) | |||
| N/total | Odds ratios (95% CI) | N/total | Odds ratios (95% CI) | |
| Unadjusted | ||||
| Low | 100/1608 | 1.34 (0.99 to 1.83) | 106/1598 | 1.46 (1.08 to 1.99) ¶ |
| Middle | 81/1589 | 1.09 (0.79 to 1.50) | 76/1597 | 1.03 (0.74 to 1.43) |
| High (referent) | 75/1596 | 1.00 | 74/1598 | 1.00 |
| P-trend § | 0.107 | 0.015 | ||
| Model 1† | ||||
| Low | 1.32 (0.91 to 1.92) | 1.84 (1.26 to 2.67) ‖ | ||
| Middle | 1.07 (0.77 to 1.51) | 1.15 (0.81 to 1.64) | ||
| High (referent) | 1.00 | 1.00 | ||
| P-trend § | 0.233 | 0.001 | ||
| Model 2‡ | ||||
| Low | 1.17 (0.79 to 1.73) | 1.80 (1.23 to 2.64) ¶ | ||
| Middle | 1.01 (0.72 to 1.43) | 1.15 (0.81 to 1.64) | ||
| High (referent) | 1.00 | 1.00 | ||
| P-trend § | 0.779 | 0.002 | ||
The tertile cut-points for each compliance measure were as follows: large vessel compliance: low, <10.7; middle, 10.7-<15.0; high (reference), ≥15.0; small vessel compliance: low, <2.8; middle, 2.8-<5.1; high (reference), ≥5.1
Model 1: adjusted for age, gender, race, cigarette smoking, hypertension, diabetes, GFR, ratio of total to HDL cholesterol, and baseline ABI.
Model 2: model 1 plus additionally adjusted for height, weight, and heart rate.
P value for linear trend across the arterial compliance tertiles was tested by entering log-transformed C1 or C2 variable in logistic regression.
p < 0.001,
p < 0.01, comparing to the reference group (highest tertile of arterial compliance or normal ABI).
Five hundred sixty eight participants had an increase in ABI > 0.15. As compared to those with no significant change in ABI, those who had an increase in ABI > 0.15 included lower proportions of Hispanics, had lower baseline ABI values, and included a higher prevalence of smokers (Table 3). There were no significant associations observed between baseline C1 or C2 measures and increase in ABI > 0.15.
When examined as a log-transformed linear variable both C1 and C2 had significant, modest beta coefficients for decline in ABI ≥ 0.15 [beta (SE): 0.08 (0.04) p < 0.05 and 0.11 (0.03) p < 0.001, respectively] and increase in ABI≥0.15 [0.13 (0.04 p<0.01 and 0.09 (.03) p<0.001, respectively].
Discussion
Findings reported here demonstrate that in cross-sectional analyses, lower ABI values are associated with significantly lower small artery compliance (C2) among men and women participants in MESA. Large artery compliance, C1, was less strongly associated with ABI in both men and women, and pairwise comparisons revealed only modest differences in C1 in the lower ABI strata. However, in pairwise comparisons, small artery compliance, C2, was significantly lower across most low ABI strata compared with the reference ABI category (ABI 1.00 – ≤1.40). In men, C2 was significantly lower across all low ABI strata. In prospective analyses, lower values of C2 at baseline were positively associated with a higher incidence of ABI decline > 0.15 at 3 years of follow up.
PAD, at least as determined by ABI, is primarily a disease of the large, muscular arteries that supply the lower extremities. It is therefore somewhat surprising that C2 appears to have a stronger cross sectional and longitudinal association to ABI than does C1. However, decreases in arterial compliance may occur in the distal arterial bed with the development of atherosclerotic disease; this may be easier to measure, or may exhibit more exaggerated differences than changes in the mechanical properties of large arteries supplying the lower extremity. In addition, some researchers have theorized that low C2 may be tightly linked to endothelial dysfunction, a well validated maker of atherosclerosis.28
To our knowledge, there are no previous reports of an independent association between C2 and a decline in ABI. As postulated above, it may be that C2 is detecting decreased arterial compliance due to endothelial dysfunction in the small vessels which may portend an increased probability of advancing lower extremity atherosclerosis. This suggests that, although the geographic distribution of atherosclerosis may vary, some systemic functional and structural changes may occur. We suspect the significant linear trend and beta coefficients for an association between lower C2 and an increase and a decrease in ABI is due to increased arterial stiffness occurring in atheroscleroitic and calcified forms of PAD progression.
Associations between arterial compliance and ABI have been investigated in two other small cross sectional studies. Duprez et al found a significant correlation between ABI and C2 (r=0.36, p < 0.02) in 43 patients with PAD.17 C2 values have been shown to associate with higher prevalence of hypertension, older age, smoking, diabetes, and with higher Framingham Risk Scores as well. Thus, it was not surprising that unadjusted positive associations between ABI and C2 were observed in the MESA cohort, but the associations we observed remained significant even after multivariate adjustement.9, 13–16
One limitation of our current study, as mentioned above, may be the geographic nature of atherosclerotic disease. PAD is primarily a disease of the large, muscular arteries of the lower extremity. However, these compliance measures were derived from pulse contour analysis of wave forms taken from the radial artery. Structural and functional changes may occur systemically in muscular arteries, suggesting that compliance derived from arterial pulse wave contour analysis from the radial artery may reflect compliance in the lower extremity arteries. However there is some doubt that radial artery derived compliance measures truly represent systemic vascular compliance.29 It would be useful to derive C1 and C2 measured in the lower extremity in patients with PAD to control for this confounder and to further elucidate potential mechanisms for the observed associations between C2 and ABI.
The method for calculating C1 and C2 from radial artery pulse wave analysis is partially proprietary and unavailable to the research team. As such, if some of our covariates, for example SBP, are included in the mathematical transformation function that is applied to the radial artery diastolic pressure decay curve, variable redundancy in the adjusted models may have occurred which could have decreased their accuracy.
In conclusion, lower arterial compliance is associated with lower ABI values in the MESA cohort. However the strength of this association is modest. Low baseline C2 values, indicating lower small arterial compliance, are associated with a higher incidence of significant decline in ABI at 3 years of follow up, even after adjustment for risk factors for PAD. Thus, radial artery tonometry derived C2 may provide additional information about the extent and potential for progression of PAD. Further study may be warranted on the strength, mechanisms of association, and possible clinical utility of these measures.
Supplementary Material
Acknowledgments
Source(s) of Funding: This research was supported by contracts N01-HC-95159 through N01-HC-95169 from the National Heart, Lung, and Blood Institute. The authors thank the other investigators, the staff, and the participants of the MESA study for their valuable contributions. A full list of participating MESA investigators and institutions can be found at http://www.mesa-nhlbi.org
Footnotes
Conflicts of interest/ Disclosures: None
References
- 1.Fung Y. Biodynamics. New York: Springer-Verlag; 1984. [Google Scholar]
- 2.Lijmer JG, Hunink MGM, van den Dungen JJAM, Loonstra J, Smit AJ. ROC analysis of noninvasive tests for peripheral arterial disease. Ultrasound in Medicine & Biology. 1996;22(4):391–398. doi: 10.1016/0301-5629(96)00036-1. [DOI] [PubMed] [Google Scholar]
- 3.Ross R. George Lyman Duff Memorial Lecture. Atherosclerosis: a problem of the biology of arterial wall cells and their interactions with blood components. Arterioscler Thromb Vasc Biol. 1981;1(5):293–311. doi: 10.1161/01.atv.1.5.293. [DOI] [PubMed] [Google Scholar]
- 4.Werns S, Walton J, Hsia H, Nabel E, Sanz M, Pitt B. Evidence of endothelial dysfunction in angiographically normal coronary arteries of patients with coronary artery disease. Circulation. 1989;79(2):287–291. doi: 10.1161/01.cir.79.2.287. [DOI] [PubMed] [Google Scholar]
- 5.Ankle Brachial Index Collaboration. Ankle Brachial Index Combined With Framingham Risk Score to Predict Cardiovascular Events and Mortality: A Meta-analysis. JAMA. 2008;300(2):197–208. doi: 10.1001/jama.300.2.197. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.McDermott MM, Guralnik JM, Tian L, Liu K, Ferrucci L, Liao Y, Sharma L, Criqui MH. Associations of Borderline and Low Normal Ankle-Brachial Index Values With Functional Decline at 5-Year Follow-Up: The WALCS (Walking and Leg Circulation Study) Journal of the American College of Cardiology. 2009;53(12):1056–1062. doi: 10.1016/j.jacc.2008.09.063. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.McDermott MM, Liu K, Greenland P, Guralnik JM, Criqui MH, Chan C, Pearce WH, Schneider JR, Ferrucci L, Celic L, Taylor LM, Vonesh E, Martin GJ, Clark E. Functional Decline in Peripheral Arterial Disease. JAMA: The Journal of the American Medical Association. 2004;292(4):453–461. doi: 10.1001/jama.292.4.453. [DOI] [PubMed] [Google Scholar]
- 8.Oliver JJ, Webb DJ. Noninvasive Assessment of Arterial Stiffness and Risk of Atherosclerotic Events. Arterioscler Thromb Vasc Biol. 2003;23(4):554–566. doi: 10.1161/01.ATV.0000060460.52916.D6. [DOI] [PubMed] [Google Scholar]
- 9.Cohn JN, Finkelstein S, McVeigh G, Morgan D, LeMay L, Robinson J, Mock J. Noninvasive Pulse Wave Analysis for the Early Detection of Vascular Disease. Hypertension. 1995;26(3):503–508. doi: 10.1161/01.hyp.26.3.503. [DOI] [PubMed] [Google Scholar]
- 10.HDI/PulseWave CR-2000 Research Cardio Vascular Proflining System Operators Manual. 2009. [Google Scholar]
- 11.Duprez DA, Jacobs DR, Lutsey PL, Bluemke DA, Brumback LC, Polak JF, Peralta CA, Greenland P, Kronmal RA. Association of Small Artery Elasticity With Incident Cardiovascular Disease in Older Adults. American Journal of Epidemiology. 174(5):528–536. doi: 10.1093/aje/kwr120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Peralta CA, Adeney KL, Shlipak MG, Jacobs D, Duprez D, Bluemke D, Polak J, Psaty B, Kestenbaum BR. Structural and Functional Vascular Alterations and Incident Hypertension in Normotensive Adults. American Journal of Epidemiology. 171(1):63–71. doi: 10.1093/aje/kwp319. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Wilson AM, O’Neal D, Nelson CL, Prior DL, Best JD, Jenkins AJ. Comparison of arterial assessments in low and high vascular disease risk groups[ast] Am J Hypertens. 2004;17(4):285–291. doi: 10.1016/j.amjhyper.2003.10.009. [DOI] [PubMed] [Google Scholar]
- 14.McVeigh G, Brennan G, Hayes R, Cohn J, Finkelstein S, Johnston D. Vascular abnormalities in non-insulin-dependent diabetes mellitus identified by arterial waveform analysis. The American Journal of Medicine. 1993;95(4):424–430. doi: 10.1016/0002-9343(93)90313-e. [DOI] [PubMed] [Google Scholar]
- 15.McVeigh G, Burns D, Finkelstein S, McDonald K, Mock J, Feske W, Carlyle P, Flack J, R G, Cohn J. Reduced vascular compliance as a marker for essential hypertension. Am J Hypertens. 1991;4:245–251. doi: 10.1093/ajh/4.3.245. [DOI] [PubMed] [Google Scholar]
- 16.McVeigh GE, Morgan DJ, Finkelstein SM, Lemay LA, Cohn JN. Vascular abnormalities associated with long-term cigarette smoking identified by arterial waveform analysis. The American Journal of Medicine. 1997;102(3):227–231. doi: 10.1016/S0002-9343(96)00454-8. [DOI] [PubMed] [Google Scholar]
- 17.Duprez D, De Buyzere M, De Bruyne L, Clement D, Cohn J. Small and large artery elasticity indices in peripheral arterial occlusive disease (PAOD) Vascular Medicine. 2001;6(4):211–214. doi: 10.1177/1358836x0100600402. [DOI] [PubMed] [Google Scholar]
- 18.Bild DE, Bluemke DA, Burke GL, Detrano R, Diez Roux AV, Folsom AR, Greenland P, Jacobs DR, Jr, Kronmal R, Liu K, Nelson JC, O’Leary D, Saad MF, Shea S, Szklo M, Tracy RP. Multi-Ethnic Study of Atherosclerosis: Objectives and Design. Am J Epidemiol. 2002;156(9):871–881. doi: 10.1093/aje/kwf113. [DOI] [PubMed] [Google Scholar]
- 19.Friedewald WT, Levy RI, Fredrickson DS. Estimation of the Concentration of Low-Density Lipoprotein Cholesterol in Plasma, Without Use of the Preparative Ultracentrifuge. Clin Chem. 1972;18(6):499–502. [PubMed] [Google Scholar]
- 20.Diagnosis and classification of diabetes mellitus. Diabetes Care. 2004;27:S5–10. doi: 10.2337/diacare.27.2007.s5. [DOI] [PubMed] [Google Scholar]
- 21.K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis. 2002;39:S1–266. [PubMed] [Google Scholar]
- 22.Criqui MH, Ninomiya JK, Wingard DL, Ji M, Fronek A. Progression of Peripheral Arterial Disease Predicts Cardiovascular Disease Morbidity and Mortality. Journal of the American College of Cardiology. 2008;52(21):1736–1742. doi: 10.1016/j.jacc.2008.07.060. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Cronenwett JL, Warner KG, Zelenock GB, Whitehouse WM, Jr, Graham LM, Lindenauer SM, Stanley JC. Intermittent Claudication: Current Results of Nonoperative Management. Arch Surg. 1984;119(4):430–436. doi: 10.1001/archsurg.1984.01390160060012. [DOI] [PubMed] [Google Scholar]
- 24.Nicoloff AD, Taylor LM, Jr, Sexton GJ, Schuff RA, Edwards JM, Yeager RA, Landry GJ, Moneta GL, Porter JM. Relationship between site of initial symptoms and subsequent progression of disease in a prospective study of atherosclerosis progression in patients receiving long-term treatment for symptomatic peripheral arterial disease. Journal of Vascular Surgery. 2002;35(1):38–47. [PubMed] [Google Scholar]
- 25.Nair N, Oka RK, Waring LD, Umoh EM, Taylor CB, Cooke JP. Vascular compliance versus flow-mediated vasodilation: correlation with cardiovascular risk factors. Vascular Medicine. 2005;10(4):275–283. doi: 10.1191/1358863x05vm633oa. [DOI] [PubMed] [Google Scholar]
- 26.McDermott MM, Liu K, Criqui MH, Ruth K, Goff D, Saad MF, Wu C, Homma S, Sharrett AR. Ankle-Brachial Index and Subclinical Cardiac and Carotid Disease: The Multi-Ethnic Study of Atherosclerosis. Am J Epidemiol. 2005;162(1):33–41. doi: 10.1093/aje/kwi167. [DOI] [PubMed] [Google Scholar]
- 27.Kennedy M, Solomon C, Manolio TA, Criqui MH, Newman AB, Polak JF, Burke GL, Enright P, Cushman M. Risk Factors for Declining Ankle-Brachial Index in Men and Women 65 Years or Older: The Cardiovascular Health Study. Arch Intern Med. 2005;165(16):1896–1902. doi: 10.1001/archinte.165.16.1896. [DOI] [PubMed] [Google Scholar]
- 28.Grey E, Bratteli C, Glasser SP, Alinder C, Finkelstein SM, Lindgren BR, Cohn JN. Reduced small artery but not large artery elasticity is an independent risk marker for cardiovascular events. American Journal of Hypertension. 2003;16(4):265–269. doi: 10.1016/s0895-7061(02)03271-5. [DOI] [PubMed] [Google Scholar]
- 29.Manning TS, Shykoff BE, Izzo JL., Jr Validity and Reliability of Diastolic Pulse Contour Analysis (Windkessel Model) in Humans. Hypertension. 2002;39(5):963–968. doi: 10.1161/01.hyp.0000016920.96457.7c. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.