A high ankle-brachial index is associated with increased aortic pulse wave velocity: the Czech post-MONICA study

Abstract

Background: Ankle brachial index (ABI) has been increasingly used in general practice to identify individuals with low ABI at high cardiovascular risk. However, there has been no consensus on the clinical significance of high ABI. The aim of our study was to compare aortic stiffness as a marker of cardiovascular risk in individuals with low (<1.0), normal (1.0–1.4), and high ABI (>1.4).

Methods: A total of 911 individuals from the Czech post-MONICA study (a randomly selected 1% representative population sample, aged 54 ± 13.5 years, 47% of men) were examined. ABI was measured using a handheld Doppler and aortic pulse wave velocity (aPWV) using the Sphygmocor device.

Results: Of the 911 individuals, 28 (3.1%) had low ABI and 23 (2.5%) high ABI. There was a U-shaped association between aPWV and ABI. aPWV was significantly higher in individuals with low and high ABI compared with the normal ABI group (11.1 ± 2.8, 8.3 ± 2.3, p < 0.001; 10.8 ± 2.5, 8.3 ± 2.3 m/s, p < 0.001, respectively). In a model adjusted for age, sex, systolic, diastolic, mean blood pressure and examiner, aPWV remained increased in both extreme ABI groups compared with the normal ABI group. In logistic regression analysis, aPWV together with glucose level, male sex, and a history of deep venous thrombosis were independent predictors of high ABI, while cholesterol was not.

Conclusion: This is the first study showing increased aortic stiffness in individuals with high ABI, presumably responsible for increased left ventricular mass described previously in this group. These findings suggest increased cardiovascular risk of high ABI individuals.

Keywords

Arterial stiffness high ankle-brachial index pulse wave velocity

Introduction

Ankle brachial index (ABI) is the ratio of the higher of ankle pressures to the higher of arm pressures. A low ABI is diagnostic for peripheral arterial disease (PAD) and is associated with increased risk of all-cause mortality, cardiovascular mortality, and total mortality as compared with normal ABI.^1,2

In contrast with the strong evidence on low ABI, the clinical significance of high ABI (ABI >1.4) is unknown. Most epidemiological studies excluded individuals with high ABI. Few studies comparing cardiovascular risk associated with high ABI have provided discrepant data. Recently, increased left ventricular mass was described in individuals with high ABI.³ We hypothesized that increased aortic pulse wave velocity (aPWV) could be responsible for increased left ventricular mass and increased cardiovascular risk in individuals with high ABI.

The aim of our study was to compare aPWV as an independent predictive factor for all-cause and cardiovascular morbidity and mortality in individuals with low (ABI <1.0), normal (ABI 1.0–1.4), and high ABI (ABI >1.4) in a random population sample.

Methods

Study population

The Czech post-MONICA (MONItoring trends and determinants in CArdiovascular disease) study is a population survey studying trends and determinants of cardiovascular risk factors in a 1% random sample of the Czech population in nine districts of the country. Methods of the Czech post-MONICA study are described elsewhere.⁴ Our study includes individuals aged >25 years from the City of Pilsen. The overall response rate in this district was 68.0%. A total of 911 patients had complete data on ABI and aPWV, which represents 98% of patients examined in the district.

Doppler ABI measurement

Appropriately sized cuffs of the mercury sphygmomanometer (Baumanometer; WA Baum, NY, USA) were placed proximal to the ankles and on the right arm. After at least 5 minutes’ resting period in the supine position, systolic blood pressure was measured in the right brachial artery, right dorsal pedal, posterior tibial artery, left dorsal pedal, and tibial artery in this order, using a pocket Doppler device with an 8 MHz probe (Dopplex multiTM; Huntleigh, Cardiff, UK). Next, systolic blood pressure was re-measured in the right brachial artery for a second time. If the difference between the first and the second brachial systolic pressure measurements was higher than 10 mmHg, all measurements were repeated. All measurements were performed by two physicians experienced in ABI measurement. ABI was calculated separately for each leg by dividing the higher of the ankle systolic pressures by the brachial systolic pressure. The lower of the two leg ABI values was used in further analysis.

Large artery properties measurement

Measurements were done using a semi-automatic Sphygmocor device (AtCor Medical, Australia) in the recumbent position. The methods of measurement were described in detail in our previous report.⁵ aPWV was calculated as the ratio of pulse wave time shift between the carotid and femoral arteries and the distance between the two sides. Consecutive registrations of the pulse waves are electrocardiogram-gated and thus, the time shift between the foot of wave at the first and second sites can be calculated. The distance between the two sites was calculated by subtracting the distance from the jugular fossa to carotid pulsation from the distance from the jugular fossa to the pulsation of the femoral artery in the groin. The average of measurements over a period of 8 s was calculated after the exclusion of extreme values.

Statistical analysis

Descriptive statistics are given as mean and standard deviation or frequency and percentage. Previous studies have demonstrated that subjects with ABI <1.0 and ≥1.4 or with incompressible arteries are at greater risk for cardiovascular disease events and mortality than subjects with normal ABI. Participants were divided into three groups on the basis of ABI according to these cut-points. Characteristics of these groups were compared using the one-way ANOVA test with Turkey’s post-hoc test for continuous variables, and the chi-squared test with Bonferroni correction for categorical variables. For continuous variables, when equal variance was violated, the Kruskal–Wallis test was used. The Fisher exact test was employed when the expected number in any cell for a categorical variable was less than 10. Variables significantly differing between the ABI categories were included into binary logistic regression analysis to evaluate factors associated with high ABI. To compare aPWV between the different ABI groups, analysis of variance and a model adjusted for age, sex, systolic, diastolic, mean blood pressure, and observer were used. A two-sided p-value <0.05 was considered to be statistically significant.

Results

Of the 911 individuals with complete data on arterial indices, 28 (3.1%) had low ABI, 860 (94.4%) normal ABI, and 23 (2.5%) high ABI. Of the 23 individuals with high ABI, five had incompressible vessels in one leg and two in both legs. Tables 1 and 2 provide population characteristics by ABI groups.

Table 1.

Descriptive statistics by ankle brachial index (ABI) groups: continuous variables

Characteristic	Low ABI n = 28 3.1%	Normal ABI n = 860 94.4%	High ABI n = 23 2.5%	Total n = 911 100%	p-value for ANOVA	p-value for low vs. normal ABI	p-value for low vs. high ABI	p-value for normal vs. high ABI
Age (years)	63.29 ± 9.24	53.45 ± 13.54	62.74 ± 7.82	53.99 ± 13.49	<0.0001	<0.0001	1.00	<0.01
Total cholesterol (mmol/l)	4.99 ± 1.0	5.17 ± 1.02	5.03 ± 0.99	5.16 ± 1.01	0.528
Triglycerides (mmol/l)	1.64 ± 0.73	1.5 ± 1.05	1.65 ± 0.83	1.51 ± 1.04	0.606
HDL-chol (mmol/l)	1.33 ± 0.27	1.43 ± 0.43	1.32 ± 0.29	1.43 ± 0.42	0.085
LDL-chol (mmol/l)	2.91 ± 0.91	2.92 ± 1.3	2.95 ± 0.79	2.92 ± 1.28	0.993
Plasma fasting glucose (mmol/l)	6.33 ± 1.99	5.32 ± 1.05	6.74 ± 3.28	5.39 ± 1.23	0.001	<0.01	1.00	<0.05
Height (cm)	167.39 ± 8.17	170.63 ± 9.25	174.04 ± 7.55	170.62 ± 9.21	<0.05	0.14	<0.05	0.13
Weight (kg)	86.57 ± 14.39	80.15 ± 16.49	90.57 ± 10.69	80.61 ± 16.41	<0.01	<0.01	0.97	<0.01
Waist circumference (cm)	104.04 ± 11.36	93.82 ± 13.66	104.35 ± 9.27	94.4 ± 13.7	<0.0001	<0.0001	0.92	<0.001
Hip circumference (cm)	112.46 ± 10.24	107.73 ± 9.37	110.00 ± 5.49	107.93 ± 9.35	<0.05	<0.05	0.314	0.606
Systolic BP (mmHg)	138.85 ± 16.92	127.08 ± 16.95	138.55 ± 17.56	127.7 ± 17.15	0.0001	<0.001	1.00	<0.01
Diastolic BP (mmHg)	67.77 ± 11.94	69.48 ± 10.78	71.77 ± 10.87	69.48 ± 10.81	0.441
Pulse pressure (mmHg)	71.08 ± 21.76	57.6 ± 19.19	66.77 ± 23.45	58.22 ± 19.53	<0.0001	<0.001	0.723	<0.05
aPWV (m/s)	11.13 ± 2.83	8.3 ± 2.28	10.79 ± 2.5	8.45 ± 2.39	0.0001	<0.0001	0.86	<0.0001

Values are mean ± SD.

aPWV, aortic pulse wave velocity; BP, blood pressure; HDL-chol, high-density lipoprotein cholesterol; LDL-chol, low-density lipoprotein cholesterol.

Table 2.

Descriptive statistics by ankle brachial index (ABI) groups: categorical variables

Characteristic	Low ABI n = 28, 3.1%	Normal ABI n = 860, 94.4%	High ABI n = 23, 2.5%	p-value for low vs. normal ABI	p-value for low vs. high ABI	p-value for normal vs. high ABI
Male sex	16 (57.1)	393 (45.8)	20 (87)	NS	NS	<0.001
CHD	7 (25)	38 (4.4)	1 (4.3)	<0.001	NS	NS
Stroke or TIA	1 (3.6)	18 (2.1)	1 (4.3)	NS	NS	NS
DVT	2 (7.1)	38 (4.4)	5 (21.7)	NS	NS	<0.05
PE	1 (3.6)	7 (0.8)	2 (8.7)	NS	NS	NS
Smokers	7 (25)	210 (24.6)	1 (4.3)	NS	NS	NS
HT	22 (78.6)	380 (44.5)	16 (69.6)	<0.01	NS	<0.05
DM	11 (39.3)	53 (6.2)	6 (26.1)	<0.001	NS	<0.01
HLP	26 (92.6)	598 (70)	16 (69.6)	<0.05	NS	NS

Values are n (%).

CHD, coronary heart disease; DM, diabetes; DVT, deep venous thrombosis; HLP, hyperlipidaemia; HT, hypertension; NS, not significant; PE, pulmonary embolism; TIA, transient ischaemic attack.

Compared with individuals with normal ABI, those with low ABI were significantly older, had higher body weight, larger waist and hip circumferences, and higher systolic blood pressure and glucose level. Prevalence of hypertension, diabetes, hyperlipidaemia, and coronary heart disease was higher in these participants compared with the normal ABI group.

Similar to individuals with low ABI, those with high ABI were older, had higher body weight, larger waist circumference, systolic blood pressure and higher glycaemia compared with participants with normal ABI. In the group with high ABI, the male sex was predominant (87% of males). A higher cardiovascular risk profile of individuals with high ABI was suggested by the higher prevalence of hypertension, diabetes, and deep venous thrombosis (DVT) as compared with those with normal ABI.

Compared with the other two groups, individuals with low ABI were more frequently treated with lipid-lowering and antihypertensive drugs (Table 3). On the other hand, the proportion of individuals with high ABI and normal ABI on lipid-lowering drugs did not differ.

Table 3.

Lipid-lowering and antihypertensive drugs by ankle brachial index (ABI) groups

Type of drug	Low ABI n = 28 3.1%	Normal ABI n = 860 94.4%	High ABI n = 23 2.5%	Total n = 911 100%	p-value for low vs. normal ABI	p-value for low vs. high ABI	p-value for normal vs. high ABI
Lipid-lowering drugs	13 (46)	145 (17)	6 (23)	164 (18)	<0.001	NS	NS
Antihypertensive medication	21 (75)	284 (33)	11 (48)	316 (35)	<0.001	NS	NS
ACEIs/ARBs	19 (68)	199 (23)	10 (43)	228 (25)	<0.001	NS	<0.05
Calcium-channel blockers	11 (39)	103 (12)	5 (22)	119 (13)	<0.01	NS	NS
Diuretics	7 (25)	90 (10)	4 (17)	101 (11)	<0.01	NS	NS
Beta-blockers	11 (40)	122 (14)	1 (4)	134 (15)	<0.01	NS	NS
Central antihypertensives	1 (4)	18 (2)	0 (0)	19 (2)	NS	NS	NS
α-blockers	0 (0)	10 (1)	1 (4)	11 (1)	NS	NS	NS

Values are n (%).

ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker.

aPWV followed a U-shaped curve with regard to ABI (Figure 1). It was significantly higher in participants with low and high ABI compared with the normal ABI group (11.1 ± 2.8 vs. 8.3 ± 2.3 m/s, p < 0.0001; 10.8 ± 2.5 vs. 8.3 ± 2.3 m/s, p < 0.0001, respectively) and did not differ between individuals with low and high ABI (11.1 ± 2.8 m/s vs. 10.8 ± 2.5 m/s, p = 0.86). In a model adjusted for age, sex, systolic, diastolic, mean blood pressure, and examiner, aPWV remained increased in both extreme ABI groups compared with the normal ABI group [9.8 ± 0.3 m/s (95% CI 9.15–10.46) vs. 8.4 ± 0.1 m/s (95% CI 8.25–8.47), p < 0.001 for low vs. normal ABI group; 9.4 ± 0.4 m/s (95% CI 8.65–10.08) vs. 8.4 ± 0.1 m/s (95% CI 8.25–8.47), p = 0.02 for high vs. normal ABI group] and no difference was found between low and high ABI groups (9.8 ± 0.3 vs. 9.4 ± 0.4 m/s, p = 0.98). In the logistic regression analysis, aPWV together with glucose level, male sex, and a history of deep venous thrombosis were independent predictors of high ABI (Table 4).

Table 4.

Parameters associated with high ankle brachial index in logistic regression analysis

	B	OR (95% CI)	SE	Wald	p-value
Plasma fasting glucose (mmol/l)	0.33	1.39 (1.05–1.86)	0.14	5.45	0.02
aPWV (m/s)	0.31	1.36 (1.16–1.64)	0.09	12.29	0.001
Sex (male)	−2.02	7.52 (1.76–36.9)	0.77	6.87	0.009
DVT	1.97	7.2 (2.3–22.4)	0.58	7.6	0.022
Age (years)	0.55	1.73 (0.78–3.82)	0.41	1.82	0.18
Weight (kg)	0.32	1.38 (0.48–3.98)	0.54	0.35	0.55
Waist circumference (cm)	−0.07	0.94 (0.29–3.02)	0.6	0.01	0.91
Systolic BP (mmHg)	0.04	1.18 (0.79–1.76)	0.2	0.68	0.41

r² = 0.2, p < 0.001.

aPWV, aortic pulse wave velocity; DVT, history of deep venous thrombosis; SE, standard error.

Discussion

ABI measurement has been increasingly used in general practice to identify individuals with low ABI at high cardiovascular risk. However, because of insufficient evidence, there is no consensus on the clinical significance of high ABI in regard to cardiovascular risk. Most large epidemiological studies have excluded these individuals. There are only a few studies showing inconsistent evidence on cardiovascular risk associated with high ABI. Increased risk of all-cause and cardiovascular mortality in individuals with ABI >1.4 was described in the Strong Heart Study in American Indians⁶ and in the Cardiovascular Health Study.⁷ Similarly, increased risk of cardiovascular mortality at ABI levels below 1.1 and above 1.3 was noted among Japanese haemodialysis patients.⁸ Contrary to these findings, the cardiovascular event rates in the Atherosclerosis Risk in Community (ARIC) study did not differ between individuals with high and normal ABI over a mean follow-up period of 12 years.⁹ There is also disagreement on the cardiovascular risk profile associated with high ABI. While the prevalence of diabetes was increased in the high ABI group in some studies,^6,7,10,11 there was no difference in others.^10,12 This discrepancy among studies could be explained by the different methods of ABI measurement used. While studies measuring ABI by Doppler showed increased risk associated with high ABI (Strong Heart Study and Cardiovascular Health Study), others measuring ABI by the oscillometric method (ARIC study)¹³ did not find increased risk associated with high ABI. As we and others have shown,¹² the difference between Doppler and oscillometric ABI increases with increasing ABI. Thus oscillometric devices underestimate high ABI. Moreover, an oscillometric device is not able to identify incompressible arteries. Thirty percent of individuals with high ABI had incompressible leg arteries in our study.

An ABI <0.9 is considered to be a sign of lower extremity PAD. There is increasing evidence showing that ABI values previously considered low normal are associated with a poor outcome similar to patients with ABI <0.9. In the Multi Ethnic Study of Atherosclerosis, individuals with ABI 0.9 – 1.10 had higher levels of subclinical atherosclerosis in the carotid and coronary arteries than those with ABI 1.10–1.30.¹⁴ Similarly, in the Strong Heart Study, total and cardiovascular mortality risk increased at ABI <1.10. These findings and increased aPWV in individuals with ABI 0.9–1.0 compared with ABI 1.0–1.4 (10.52 ± 2.9 m/s vs. 8.24 ± 2.14 m/s, p < 0.0001) made us choose ABI = 1.0 as the cut-off for the low ABI group.

We used ABI >1.4 as the cut-off value for the high ABI group. There were 59 individuals with an ABI between 1.3 and 1.4. When setting the cut-off value for high ABI to 1.3, the difference in aPWV between the low and high ABI groups became significant (11.1 ± 2.8 m/s vs. 9.2 ± 2.5 m/s, p < 0.001), while the difference between the normal and high ABI groups remained significant (8.3 ± 2.3 m/s vs. 9.2 ± 2.5 m/s, p < 0.01). After adjustment for age, sex, systolic, diastolic, mean blood pressure, and examiner, the difference between normal and high ABI was no longer significant [8.4 ± 0.2 m/s (95% CI 8.0–8.8) vs. 8.6 ± 0.3 m/s (95% CI 8–9.1)]. This means that in individuals with ABI between 1.3 and 1.4, aPWV is not significantly increased.

The observed prevalence of high ABI in our population sample was 2.5%. It is higher than the 1.15% and 1.2% prevalence of ABI >1.4 in the Cardiovascular Health Study and ARIC study, respectively, but lower than the 9.2% prevalence reported in the Strong Heart Study. An even higher prevalence was reported in individuals with chronic renal failure (23.7%) or on dialysis (41.7%).¹⁵ The prevalence of high ABI observed in our random population sample of the Czech population was comparable with that of low ABI. This suggests that high ABI is not a rare condition in the population.

An abnormally increased ABI is widely believed to be associated with medial arterial calcification caused by calcification of the arterial media and the internal elastic membrane of muscular arteries.¹⁶ Vascular calcification is the consequence of dysregulation between promotion and inhibition of calcification often seen in chronic kidney disease, diabetes mellitus, atherosclerosis and aging.¹⁷ Thus a high ABI is not exclusively associated with diabetes as commonly believed. The prevalence of diabetes in the high ABI group in our study (25%) is similar to the 28% prevalence in the Multi-Ethnic Study of Atherosclerosis and the 25% prevalence in the Cardiovascular Health Study. The lower prevalence of diabetes in our and other studies may be due to the definition of diabetes based on use of antidiabetic drugs and fasting glucose level. Definitions based also on the oral glucose tolerance test or glycated haemoglobin level may have resulted in an increased prevalence of diabetes in the high ABI group.

Another important finding is that cholesterol level is not associated with high ABI. The prevalence of hyperlipidaemia was increased in the low ABI group compared with the normal ABI group, and no difference was found between the normal and high ABI groups. There was no difference in cholesterol levels between the groups, which was most likely due to the high rates of use of lipid-lowering drugs in low ABI individuals (Table 3). This might be yet another proof of the concept that high ABI is linked to stiffness and not to atherosclerosis.

Increased aortic stiffness increases systolic and pulse pressure while decreasing diastolic blood pressure. In our and other studies,^3,7 high ABI individuals had slightly increased systolic and pulse pressure, while no difference was seen in diastolic blood pressure compared with the normal ABI group. This could be caused by the frequent use of antihypertensive therapy in the high and low ABI groups in our study. The increased aortic stiffness persisting elevated after blood pressure control is a sign of intrinsic vasculopathy as noted by Guerin et al.¹⁸ This suggests that the increased aortic stiffness in high ABI is due to structural changes of the aorta, presumably calcifications, seen in the muscular arteries of these individuals.

Lately, there has been increasing evidence on the association between atherosclerosis and DVT.¹⁹ Conventional cardiovascular risk factors have been shown to increase the risk of DVT. The proposed mechanism behind this association is arterial and venous endothelial dysfunction. In our study, we have shown increased prevalence of DVT in the high ABI group compared with the normal ABI group. The higher prevalence of hypertension, diabetes and obesity in these individuals can cause endothelial dysfunction leading to higher risk of DVT. The non-significant difference between the low and normal ABI groups is probably caused by lack of power to detect this difference due to the small sample size. Another explanation for the increased rates of DVT in high ABI individuals may be increased lower extremity arterial stiffness due to medial arterial calcification. Increased lower extremity arterial stiffness has been shown to decrease arterial flow volume in the lower extremities of diabetic patients.²⁰ A decreased flow volume in the arterial system presumably leads to blood stasis in the venous system, an abnormality of Virchow’s triad associated with thrombus formation.

To the best of our knowledge, this is the first study showing increased aPWV in individuals with ABI >1.4. Increased arterial stiffness in patients with PAD has been reported previously. In the Health ABC study, the prevalence of ABI <0.9 increased with increasing quartiles of aPWV.²¹ In the Rotterdam study, the presence of PAD was associated with a significantly increased aPWV.²² Matsumae reported higher aPWV in non-diabetic haemodialysis patients with PAD compared with non-diabetic haemodialysis individuals without PAD.²³ In our study, we found a U-shaped association between ABI and aPWV. In agreement with other studies, we observed increased aPWV in PAD individuals. Moreover, we observed increased aPWV in the high ABI group which was comparable with those with low ABI. Increased aPWV has been shown to be an independent predictive factor for all-cause and cardiovascular mortality, cardiovascular disease, fatal and non-fatal coronary events and fatal strokes in patients with various levels of cardiovascular risk.²⁴ Increased aortic stiffness through increased left ventricular afterload leads to left ventricular hypertrophy. Increased aPWV can explain the atherosclerosis-independent increase in left ventricular mass recently described in high ABI individuals in Multi-Ethnic Study of Atherosclerosis.

To summarize, we have shown increased aortic pulse wave velocity in individuals with high ABI. The increased aPWV in these individuals indicates increased cardiovascular risk. Prospective studies using the Doppler method of ABI measurement are needed to confirm the increased cardiovascular risk associated with high ABI.

Footnotes

Funding

The study was supported by the Internal Grant Agency of the Ministry of Health of the Czech Republic (grant number NR/9389-3).

Conflict of interest

None

References

Heald

Fowkes

Murray

Price

. Risk of mortality and cardiovascular disease associated with the ankle-brachial index: systematic review. Atherosclerosis 2006; 189(1): 61–69.

Wautrecht

Dramaix

De Backer

Kornitzer

. Ankle brachial index is an independent predictor of all-cause and coronary mortality in healthy middle-aged working Belgian men after eighteen year follow-up. Eur J Cardiovasc Prev Rehabil 2006; 13(Suppl 1): S26–S27.

Katz

Peralta

de Boer

Allison

Bluemke

. A high ankle brachial index is associated with greater left ventricular mass MESA (Multi-Ethnic Study of Atherosclerosis). J Am Coll Cardiol 2010; 55(4): 342–349.

Cifkova

Skodova

Lanska

Adamkova

Novozamska

Jozifova

. Prevalence, awareness, treatment, and control of hypertension in the Czech Republic. Results of two nationwide cross-sectional surveys in 1997/1998 and 2000/2001; Czech Post-MONICA Study. J Hum Hypertens 2004; 18(8): 571–579.

Filipovsky

Ticha

Cifkova

Lanska

Stastna

Roucka

. Large artery stiffness and pulse wave reflection: results of a population-based study. Blood Press 2005; 14(1): 45–52.

Resnick

Lindsay

McDermott

Devereux

Jones

Fabsitz

. Relationship of high and low ankle brachial index to all-cause and cardiovascular disease mortality: the Strong Heart Study. Circulation 2004; 109(6): 733–739.

O'Hare

Katz

Shlipak

Cushman

Newman

. Mortality and cardiovascular risk across the ankle-arm index spectrum: results from the Cardiovascular Health Study. Circulation 2006; 113(3): 388–393.

Ono

Tsuchida

Kawai

Matsuo

Wakamatsu

Maezawa

. Ankle-brachial blood pressure index predicts all-cause and cardiovascular mortality in hemodialysis patients. J Am Soc Nephrol 2003; 14(6): 1591–1598.

Wattanakit

Folsom

Duprez

Weatherley

Hirsch

. Clinical significance of a high ankle-brachial index: insights from the Atherosclerosis Risk in Communities (ARIC) Study. Atherosclerosis 2007; 190(2): 459–464.

10.

Aboyans

Denenberg

Natarajan

Criqui

. The association between elevated ankle systolic pressures and peripheral occlusive arterial disease in diabetic and nondiabetic subjects. J Vasc Surg 2008; 48(5): 1197–1203.

11.

Allison

Hiatt

Hirsch

Coll

Criqui

. A high ankle-brachial index is associated with increased cardiovascular disease morbidity and lower quality of life. J Am Coll Cardiol 2008; 51(13): 1292–1298.

12.

Pan

Staessen

Wang

. Comparison of three measures of the ankle-brachial blood pressure index in a general population. Hypertens Res 2007; 30(6): 555–561.

13.

Tabara

Igase

Kido

Ochi

Miki

Kohara

. Composition of lower extremity in relation to a high ankle-brachial index. J Hypertens 2009; 27(1): 167–173.

14.

McDermott

Liu

Criqui

Ruth

Goff

Saad

. Ankle-brachial index and subclinical cardiac and carotid disease: the multi-ethnic study of atherosclerosis. Am J Epidemiol 2005; 162(1): 33–41.

15.

Leskinen

Salenius

Lehtimaki

Huhtala

Saha

. The prevalence of peripheral arterial disease and medial arterial calcification in patients with chronic renal failure: requirements for diagnostics. Am J Kidney Dis 2002; 40(3): 472–479.

16.

Micheletti

Fishbein

Currier

Fishbein

. Monckeberg sclerosis revisited: a clarification of the histologic definition of Monckeberg sclerosis. Arch Pathol Lab Med 2008; 132(1): 43–47.

17.

Johnson

Leopold

Loscalzo

. Vascular calcification: pathobiological mechanisms and clinical implications. Circ Res 2006; 99(10): 1044–1059.

18.

Guerin

Blacher

Pannier

Marchais

Safar

London

. Impact of aortic stiffness attenuation on survival of patients in end-stage renal failure. Circulation 2001; 103(7): 987–992.

19.

Gresele

Momi

Migliacci

. Endothelium, venous thromboembolism and ischaemic cardiovascular events. Thromb Haemost 2010; 103(1): 56–61.

20.

Suzuki

Kashiwagi

Nishio

Egawa

Shimizu

Maegawa

. Increased arterial wall stiffness limits flow volume in the lower extremities in type 2 diabetic patients. Diabetes Care 2001; 24(12): 2107–2114.

21.

Sutton-Tyrrell

Najjar

Boudreau

Venkitachalam

Kupelian

Simonsick

. Elevated aortic pulse wave velocity, a marker of arterial stiffness, predicts cardiovascular events in well-functioning older adults. Circulation 2005; 111(25): 3384–3390.

22.

van Popele

Grobbee

Bots

Asmar

Topouchian

Reneman

. Association between arterial stiffness and atherosclerosis: the Rotterdam Study. Stroke 2001; 32(2): 454–460.

23.

Matsumae

Abe

Murakami

Ishihara

Ueda

Saito

. Determinants of arterial wall stiffness and peripheral artery occlusive disease in nondiabetic hemodialysis patients. Hypertens Res 2007; 30(5): 377–385.

24.

Laurent

Boutouyrie

. Arterial stiffness: a new surrogate end point for cardiovascular disease? J Nephrol 2007; 20(Suppl 12): S45–S50.