- Email: [email protected]
Adiponectin and Arterial Stiffness
AJH
2005; 18:1543–1548
Blood Vessels
Adiponectin and Arterial Stiffness Azra Mahmud and John Feely Background: Adiponectin, an anti-inflammatory adipocytokine, is reduced in hypertension, diabetes, and coronary artery disease (CAD). Arterial stiffness, as aortic pulse wave velocity (PWV) in hypertension and diabetes, and as augmentation index (AIx) in CAD, is independently associated with cardiovascular mortality. We explored the relationship between adiponectin and arterial stiffness in essential hypertension. Methods: Seventy-six untreated patients, 34 women, aged 47 ⫾ 1 years, mean ⫾ SEM with essential hypertension, had blood pressure (BP), carotid–femoral PWV, AIx plasma adiponectin, and proinflammatory cytokine C-reactive protein (CRP) measured using ELISA technique after an overnight fast. Results were analyzed using univariate and multiple logistic regression analysis. Results: There was a significant positive relationship between log adiponectin and AIx (r ⫽ 0.33, P ⬍ .005) and plasma HDL-cholesterol (r ⫽ 0.40, P ⬍ .001). In contrast there were significant negative relationships with PWV (r
⫽ ⫺0.24, P ⬍ .05), transit time (r ⫽ ⫺0.37, P ⬍ .001), and pulse pressure amplification (r ⫽ ⫺0.30, P ⬍ .005) in addition to waist circumference, waist-to-hip ratio, height, and weight. In a stepwise regression model, the independent predictors of AIx were heart rate, height, mean arterial pressure, age, and gender (R2 ⫽ 0.69, P ⬍ .0001) with no contribution from adiponectin. However, for PWV (R2 ⫽ 0.59, P ⬍ .0001) the independent determinants were mean arterial pressure, age, and adiponectin. Conclusions: These results show a divergent relationship between adiponectin and arterial stiffness, negative for PWV, and positive for wave reflection (AIx). Anthropomorphic factors, particularly height, weight, and heart rate may influence the relationship to the latter. Adiponectin is an independent predictor of aortic PWV but not of AIx. Am J Hypertens 2005;18:1543–1548 © 2005 American Journal of Hypertension, Ltd. Key Words: Adiponectin, arterial stiffness, hypertension.
diponectin has emerged in the past few years as an important adipocytokine modulating insulin resistance with anti-inflammatory and antiatherogenic properties.1 It is expressed and excreted only in the adipose tissue and plasma levels are relatively high, representing approximately 0.01% of total plasma proteins and concentrations exceed those of many other common hormones. Adiponectin levels are related positively to HDLcholesterol but inversely to insulin1 and its levels are reduced in obesity and type 2 diabetes.2 Insulin has inhibitory effects on the levels of adiponectin.1 More recently, adiponectin has been shown to possess anti-inflammatory properties and may modulate the process of atherogenesis,1 suggesting that adiponectin deficiency associated with obesity, metabolic syndrome, and type 2 diabetes may contribute to accelerated atherosclerosis in such conditions. Atherosclerotic changes in large arteries make an important contribution to pathogenesis of cardiovascular disease. Increased aortic stiffness measured as carotid–
A
femoral pulse wave velocity (PWV) is associated with increased risk of cardiovascular events in the hypertensive3 and diabetic4 populations. Augmentation index (AIx), a composite of wave reflection from medium-sized muscular arteries, left ventricular ejection, and PWV, is related to the development of coronary artery disease (CAD).5 The PWV is significantly correlated with plasma glucose levels in subjects with impaired glucose tolerance6 and there is a significant relationship between plasma insulin levels and aortic stiffness in the general population7 and in subjects with metabolic syndrome.8 There is also a significant association between the changes in brachial ankle PWV and adiponectin in type 2 diabetes patients after treatment with pioglitazone, which significantly reduced both brachial ankle PWV and increased adiponectin concentrations.9 However, there is no information on the relationship between AIx and adiponectin in hypertension. Therefore, the aim of our study was to explore the relationship between adiponectin and arterial stiffness determined by both PWV, a classic marker of aortic stiffness
Received March 14, 2005. First decision June 9, 2005. Accepted June 10, 2005. From the Department of Therapeutics and Hypertension Clinic, Trinity Centre for Health Sciences, St. James’s Hospital, Dublin, Ireland.
Address correspondence and reprint requests to Professor John Feely, Department of Therapeutics and Hypertension Clinic, Trinity Centre for Health Sciences, St. James’s Hospital, Dublin 8, Ireland; e-mail: [email protected]
© 2005 by the American Journal of Hypertension, Ltd. Published by Elsevier Inc.
0895-7061/05/$30.00 doi:10.1016/j.amjhyper.2005.06.014
1544
ADIPONECTIN AND ARTERIAL STIFFNESS
and AIx, reflecting the stiffness of the medium-sized muscular arteries, in a never-treated hypertensive population.
Methods Patients and Methods We recruited 76 never-treated hypertensive subjects with diagnosis based on three outpatient measures of blood pressure (BP) greater than 140/90 mm Hg and confirmed by daytime ambulatory BP monitoring (⬎135/80 mm Hg). The mean (⫾SEM) age was 47 ⫾ 1 years (range 38 to 64 years), 34 were women and 26% of the patients were smokers. None of these subjects had received treatment for hypertension or any other medications that may influence BP or arterial stiffness. None of the patients had secondary causes of hypertension, CAD, renal disease, liver disorders, or diabetes. Patients were studied in the morning after an overnight fast having refrained from caffeinecontaining beverages, alcohol, and smoking in the previous 12 h. The subjects gave informed consent and the study had institutional ethics committee approval. Laboratory Measurements Height, weight, waist, and hip measurements were recorded for each patient. Venous samples were drawn into EDTA tubes, centrifuged (4°C, 2500 rpm) for 10 min and supernatant stored at ⫺80°C. Plasma adiponectin, C-reactive protein (CRP), tumor necrosis factor-␣ (TNF-␣), and interleukin-6 (IL-6) levels were measured by enzymelinked immunosorbent assay (ELISA) (R & D Systems Ltd., Abingdon, Oxford, UK) in duplicates. In addition, the plasma levels of total cholesterol, LDL, HDL, glucose, and creatinine were measured by standard methods and the creatinine clearance calculated. Hemodynamic Measurements After a supine rest of 15 min, BP and heart rate was measured with an automated oscillometric device (Omron model HEM 705 CP, Omron Corporation, Tokyo, Japan) in the right arm after which AIx and carotid femoral PWV were measured as previously described.3 Arterial Pulse Wave Analysis Radial applanation tonometry was used to acquire the aortic pressure waveform noninvasively using the generalized transfer fraction (Sphygmocor, version 7.1, Sydney, Australia). The aortic pressure waveform was used to derive the aortic BP and the AIx (difference in height of the first and second systolic peak in the aorta and aortic pulse pressure). Time to reflected wave (TR) was measured from the foot of the wave to the inflection point on the waveform and left ventricular ejection was measured from the foot of the wave to the diastolic incisura. The pulse pressure amplification was calculated as the difference between brachial and aortic pulse pressure.
AJH–December 2005–VOL. 18, NO. 12
Pulse Wave Velocity Carotid–femoral PWV was measured with an automated system (Artech Medical, Coléon, France) using the footto-foot method. The carotid and femoral waveforms were acquired simultaneously with two pressure-sensitive transducers and the transit time of the pulse calculated by the system software. The distance between the two arterial sites was measured on the body using a tape measure and PWV calculated as the distance divided by time (meters per seconds). At least 12 successive readings were used for analysis to cover a complete respiratory cycle. Statistical Analysis Statistical analysis performed with JMP version 5.0 (SAS for Windows, Cary, NC, USA). As PWV adiponectin and the inflammatory cytokines were not normally distributed they were log transformed for further analysis. The coefficient of variation of all the primary measures PWV, AIx, adiponectin, and inflammatory markers was under 5%. Relationship between adiponectin, PWV, AIx, and other parameters was analyzed using nonparametric methods (Spearman correlations). To study whether there was an independent relationship between PWV and AIx with adiponectin, stepwise regression analysis was used. In the first model, we took PWV as the dependent variable and age, mean arterial pressure, gender, body mass index (BMI), waist/hip ratio, heart rate, smoking status, adiponectin and CRP, plasma glucose, and lipoproteins as independent variables. In the second model, with the same independent variables, AIx was analyzed as the dependent variable. Results are expressed as mean ⫾ SEM; P ⬍ .05 considered significant.
Results Baseline data for the patient population is summarized in Tables 1 and 2. The relationship between adiponectin and measures of arterial stiffness and other hemodynamic variables and anthropological parameters is shown in Table 3. Adiponectin was inversely related to subject’s height, weight, waist, waist-to-hip ratio, but not BMI. There was also an inverse correlation between PWV, TR, pulse pressure amplification, and adiponectin. In contrast the significant relationship with AIx was positive. There was no relationship between adiponectin and brachial or aortic BP, with the brachial diastolic BP being closest to achieving statistical significance (P ⫽ .07). Although the relationship with heart rate was not statistically significant there was a positive relationship with left ventricular ejection duration. Adiponectin was positively correlated with HDL-cholesterol and negatively with glucose and triglycerides, but not with total cholesterol. There was no significant relationship between adiponectin levels and CRP (r ⫽ ⫺0.08), TNF-␣ (r ⫽ ⫺0.10), and IL-6 (r ⫽ ⫺0.13). The relationship between PWV and other parameters is given in Table
AJH–December 2005–VOL. 18, NO. 12
ADIPONECTIN AND ARTERIAL STIFFNESS
Table 1. Baseline characteristics of the patient population (n ⫽ 76) Parameter Age (yr) Gender (M/F) Height (cm) Weight (kg) Body mass index (kg/m2) Waist (cm) Hip (cm) Waist/hip ratio Brachial systolic BP (mm Hg) Brachial diastolic BP (mm Hg) Mean arterial pressure (mm Hg) Brachial pulse pressure (mm Hg) Heart rate (min⫺1) Aortic systolic BP (mm Hg) Aortic diastolic BP (mm Hg) Aortic pulse pressure (mm Hg) Total cholesterol (mmol/L) HDL-cholesterol (mmol/L) LDL-cholesterol (mmol/L) Triglycerides (mmol/L) Glucose (mmol/L) Serum creatinine (mmol/L) Creatinine clearance (mL/min)
Value (ⴞ mean SEM) 47 ⫾ 1 (44/32) 171 ⫾ 0.1 88.6 ⫾ 2 30 ⫾ 0.5 96.5 ⫾ 2 105 ⫾ 1 0.92 ⫾ 0.01 152 ⫾ 2 92 ⫾ 1 114 ⫾ 2 62 ⫾ 2 70 ⫾ 1 141 ⫾ 2 93 ⫾ 1 48 ⫾ 1.6 5.17 ⫾ 0.1 1.27 ⫾ 0.03 2.95 ⫾ 0.1 1.76 ⫾ 0.1 5.6 ⫾ 0.1 88.6 ⫾ 2 109 ⫾ 4
1545
Table 3. Univariate relationship between plasma adiponectin concentrations and hemodynamic, biochemical, and anthropomorphic variables (n ⫽ 76) Variables Pulse wave velocity (m/sec) Augmentation index (%) TR (msec) Pulse pressure amplification (mm Hg) Left ventricular ejection duration (msec) Weight (kg) Height (cm) Waist (cm) Waist/hip ratio Plasma glucose (mmol/L) Plasma triglycerides (mmol/L) Plasma HDL (mmol/L)
Correlation Coefficient
P
⫺0.24 0.33 ⫺0.37
⬍.05 ⬍.01 ⬍.001
⫺0.30
⬍.01
0.28 ⫺0.32 ⫺0.55 ⫺0.23 ⫺0.42
⬍.01 ⬍.01 ⬍.001 ⬍.05 ⬍.001
⫺0.23
⬍.05
⫺0.25 0.40
⬍.05 ⬍.0001
arterial pressure, age, and gender (R2 ⫽ 0.69, P ⬍ .0001) with no contribution from adiponectin.
Discussion 4. There was no difference in plasma adiponectin levels, PWV, and AIx between smokers and nonsmokers. The PWV was significantly and positively related to age, BP, AIx, and plasma glucose and inversely to TR and HDLcholesterol with no relationship with heart rate, BMI, waist, and hip measurements or LDL and total cholesterol. In a stepwise multiple regression model, plasma adiponectin levels independently predicted PWV, when adjusted for age, mean arterial pressure, gender, height, weight, waist-to-hip ratio, conventional cardiovascular risk factors, and CRP with age and mean arterial pressure the other independent predictors (R2 ⫽ 0.62, P ⬍ .0001; Table 5). However, when adjusted for the subject’s anthropomorphic measurements, cardiovascular risk factors and CRP, AIx was predicted by heart rate, height, mean
Our results show a significant inverse relationship between plasma adiponectin levels and PWV in hypertensive subjects. The finding of a negative relationship with PWV is in keeping with previous observations relating both PWV and adiponectin to cardiovascular risk. An increased PWV is associated with cardiovascular events in the hypertensive3 and diabetic4 populations and in subjects with endstage renal disease (ESRD).10 Low plasma adiponectin concentrations may predict the risk of acute coronary syndrome,11 but not of restenosis after coronary stenting.12 Adiponectin modulates endothelial adhesion molecules13 and has been found in the subendothelial space of the subcarotid arteries injured by a catheter and in atheroscle-
Table 4. Univariate correlations for pulse wave velocity Table 2. Baseline parameters for arterial stiffness, adiponectin, and markers of inflammation C-reactive protein, tumor necrosis factor (TNF)-␣, and interleukin (IL-6) Pulse wave velocity (m/sec) Augmentation index (%) Transit time (msec) Pulse pressure amplification (mm Hg) Adiponectin (g/mL) C-reactive protein (mg/L) TNF-␣ (pg/mL) IL-6 (pg/mL)
10.4 ⫾ 0.2 27 ⫾ 1.3 137 ⫾ 1.6 12.5 8.2 0.36 1.0 1.7
⫾ ⫾ ⫾ ⫾ ⫾
0.7 0.4 0.05 0.04 0.2
Variables Age Brachial systolic BP (mm Hg) Brachial diastolic BP (mm Hg) Mean arterial pressure (mm Hg) Augmentation index (%) Transit time (msec) Plasma glucose (mmol/L) Plasma HDL (mmol/L)
Correlation Coefficient
P
0.56
⬍.0001
0.74
⬍.0001
0.50
⬍.001
0.65 0.26 ⫺0.26 0.30 ⫺0.26
⬍.0001 ⬍.05 ⬍.05 ⬍.05 ⬍.05
1546
ADIPONECTIN AND ARTERIAL STIFFNESS
AJH–December 2005–VOL. 18, NO. 12
Table 5. Multiple regression models with pulse wave velocity (PWV R2 ⫽ 0.59, P ⬍ .001) and augmentation index (Alx R2 ⫽ 0.69, P ⬍ .001) as the dependent variable Parameters for PWV Mean arterial pressure (mm Hg) Age (yr) Adiponectin (g/mL) Parameters for AIx Heart rate (min⫺1) Height (cm) Mean arterial pressure (mm Hg) Age (yr) Gender (female)
R2

SE
P
0.41 0.53 0.57
0.06 0.07 ⫺0.10
0.01 0.01 0.04
⬍.0001 ⬍.0001 ⬍.01
0.21 0.42 0.62 0.66 0.71
⫺0.51 ⫺0.28 0.33 0.29 3.0
0.06 0.11 0.06 0.07 1.1
⬍.0001 ⬍.01 ⬍.0001 ⬍.0001 ⬍.01
rotic lesions within the endothelium14 and may prevent vascular restenosis.15 Circulating adiponectin concentrations are protective against CAD16 and also predictive of subsequent cardiovascular events in patient with ESRD.17 Aortic distensibility measured by ultrasound was lower in patients with chronic renal failure, suggestive of increased stiffness, but there was no relationship between distensibility and adiponectin in either controls or the patients with renal disease.18 However, in diabetic patients where adiponectin levels are reduced2 PWV is increased. Reduced adiponectin levels are associated with increased future coronary heart disease events in men with type 2 diabetes.19 In subjects with type 2 diabetes there is a significant correlation between changes in adiponectin concentrations and changes in the ankle brachial pulse wave estimate of PWV9 after treatment with pioglitazone. Studies looking at adiponectin concentrations in patients with hypertension have produced conflicting results. Adamczak et al20 reported lower concentrations, whereas Mallamaci et al21 showed higher adiponectin concentrations in hypertensive patients. Furuhashi et al22 found reduced adiponectin concentration in a group (40%) of hypertensive patients who had insulin resistance determined by using a euglycemic hyperinsulinemic glucose clamp technique. It has recently been shown23 that essential hypertensive patients with a nondipping pattern show more prominent insulin resistance and lower adiponectin levels compared to dippers. No correlation was found between adiponectin and systolic/diastolic (clinic or 24-h ambulatory) BP levels,17,21–23 although one study20 did show a negative correlation with systolic, diastolic, and mean arterial pressure. Whether the relationship between adiponectin and PWV extends to a normotensive population or a “threshold” level of pressure is required to lower adiponectin levels was not studied. Our data did not provide strong evidence of an association between adiponectin concentration and brachial or aortic BPs. This study also provides an interesting dichotomy in the relationship between indices of arterial stiffness, both independent cardiovascular risk factors, and the concentration of the adipocytokine adiponectin, which itself may play a significant role in reducing cardiovascular risk. In
contrast to PWV, the relationship between arterial stiffness as assessed by AIx or wave reflection, and adiponectin is a positive one (Figure 1). How adiponectin and PWV are related is not immediately clear and may possibly be attributable to other metabolic and vasoactive factors not measured in the study. For example, adipocytes also secrete TNF, leptin, resistin, and a number of other growth factors.1 Because there is evidence that systemic inflammation, as represented by C-reactive protein concentration, is related positively to PWV,24 one may speculate that the relationship is due to a reduced anti-inflammatory effect as a consequence of lower concentrations of adiponectin. However, we did not detect any relationship between adiponectin concentration or other markers of inflammation—TNF, IL-6, and CRP. Furthermore, adiponectin independently predicted PWV, even when CRP is added in the model. Unfortunately we have not measured insulin resistance, which is common in hypertension, but there was a negative relationship between adiponectin and plasma glucose suggesting insulin resistance, which we know increases PWV, and this may be one mechanism. Adiponectin also binds to collagen but to separate the anti-inflammatory and insulin resistance components of adiponectin it may be necessary to measure subtypes, trimers, and the carboxy-terminal globular domain that activates AMP-activated protein kinase and the hexamer and high molecular isoforms that activate nuclear factor- signaling pathways.25 We did not observe any difference between smokers and nonsmokers in relation to plasma adiponectin levels, PWV, or AIx in this study, although we have reported previously a significantly higher AIx in a young healthy normotensive smoking cohort compared to the nonsmoking group.26 However, this may be due to the small number in our study of middle-aged hypertenisves and as suggested earlier,26 an age-related effect of smoking on BP. Also the effect of smoking on arterial stiffening may be quite different in healthy compliant arteries compared to hypertensive patients with established arterial stiffening. An examination of the determinants of the two parameters of stiffness that we measured may provide some insight as to their divergent relationship with adiponectin. As previously reported3 and seen in this study age and BP
AJH–December 2005–VOL. 18, NO. 12
ADIPONECTIN AND ARTERIAL STIFFNESS
1547
uals have higher adiponectin and higher AIx, the latter in part due to the shorter distance for wave travel to reflection sites. It is well established31 that taller individuals have not only lower AIx but also increased pulse pressure amplification. Therefore, the negative relationship to amplification in this study suggests that these hemodynamic influences may be determinants for the positive relationship between AIx and adiponectin. In multivariate analysis adiponectin is not an independent determinant of AIx (in contrast to PWV) and its relationship is one of association with the other, particularly with anthropomorphic factors determining AIx. This study also emphasizes the need to have complimentary but independent measures of arterial stiffness.
References 1. 2.
3.
4.
5. FIG. 1. Relationship between plasma adiponectin and pulse wave velocity (top) and augmentation index (bottom) in patients with essential hypertension (n ⫽ 76).
are the major determinants of PWV. Both age and BP also contribute to AIx, which is a composite of the extent and site of wave reflection, the rate of wave propagation— largely PWV and ventricular ejection. However, arterial wave reflection is influenced particularly by anthropomorphic (eg, height, gender) factors in addition to hemodynamic factors including heart rate and site of the reflected wave. In this study heart rate and height were important independent determinants of AIx (Table 5), and although the negative correlation between adiponectin and heart rate was not significant, it was positively correlated with ejection duration. Healthy young men (college students) with high normal BP have lower serum adiponectin and higher heart rates than men with optimal BP, even after adjustment for BMI.27 Heart rate is inversely and independently related to insulin sensitivity.28 Sympathetic overactivity may be a common preceding feature of higher heart rate and insulin resistance.27,29 Relative sympathetic activation, as derived from heart rate variation, was associated with low adiponectin in patients with non-insulin-dependent diabetes mellitus.30. It is clear from the anthropomorphic measurements that the strongest correlation with adiponectin concentrations is body height. Shorter individ-
6.
7.
8.
9.
10.
11.
12.
13.
Gil-Campos M, Canete R, Gil A: Adiponectin, the missing link in insulin resistance and obesity. Clin Nutr 2004;23:963–974. Hotta K, Funahashi T, Arita Y: Plasma concentrations of a novel, adipose-specific protein, adiponectin, in type 2 diabetic patients. Arterioscl Thromb Vasc Biol 2000;20:1595–1599. Safar ME, Henry O, Meaume S: Aortic pulse wave velocity: an independent marker of cardiovascular risk. Am J Ger Cardiol 2002; 11:295–298. Cruickshank K, Riste L, Anderson SG, Wright JS, Dunn G, Gosling RG: Aortic pulse wave velocity and its relationship to mortality in diabetes and glucose intolerance: an integrated index of vascular function? Circulation 2002;106:2085–2090. Weber T, Auer J, O’Rourke MF, Kvas E, Lassnig E, Berent R, Eber B: Arterial stiffness, wave reflections and the risk of coronary artery disease. Circulation 2004;109:184 –189. Ohnishi H, Saitoh S, Takagi S: Pulse wave velocity as an indicator of atherosclerosis in impaired fasting glucose: the Tanno and Sobetsu study. Diabetic Care 2003;26:437– 440. Hansen TW, Jeppessen J, Rasmussen S, Ibsen H, Torp-Pedersen C: Relation between insulin and aortic stiffness: a population based study. J Hum Hypertens 2004;18:1–7. Choi KM, Lee KW, Soe JA, Oh JH, Kim SG, Kim NH, Choi DS, Baik SH: Relationship between brachial ankle pulse wave velocity and cardiovascular risk factors of the metabolic syndrome. Diabetes Res Clin Prac 2004;66:57– 61. Satoh N, Ogawa Y, Usui T, Tagami T, Kono S, Uesugi H, Sugiyama H, Sugawara A, Yamada K, Shimatsu A, Kuzuya H, Nakao K: Antiatherogenic effect of pioglitazone in type 2 diabetic patients irrespective of the responsiveness to its antidiabetic effect. Diabetes Care 2003;26:2493–2499. Blacher J, London GM, Safar B, Mourad J-J: Influence of age and end-stage renal disease on the stiffness of carotid wall material in hypertension. J Hypertens 1999;17:237–244. Nakamura Y, Shimada K, Fukuda D, Shimada Y, Ehara S, Hirose M, Kataoka T, Kamimori K, Shimodozono S, Kobayashi Y, Yoshiyama M, Takeuchi K, Yoshikawa J: Implication of plasma concentrations of adiponectin in patients with coronary artery disease. Heart 2004:90:528 –533. Shimada K, Miyauchi K, Mokuno H, Miyazaki T, Seki E, Watanabe Y, Iwama Y, Shigekiyo M, Matsumoto M, Okazaki S, Tanimoto K, Kawamura M, Suzuki H, Kurata T, Sato H, Daida H: Predictive value of the adipocyte-derived plasma protein adiponectin for restenosis after elective coronary stenting. Jpn Heart J 2002;43:85–91. Ouchi N, Kihara S, Arita Y: Novel modulator for endothelial adhesion molecules: adipocyte-derived plasma protein adiponectin. Circulation 1999;100:2473–2476.
1548
ADIPONECTIN AND ARTERIAL STIFFNESS
14. Okamoto Y, Arita Y, Nishida M: An adipocyte-derived plasma protein, adiponectin, adheres to injured vascular walls. Horm Metab Res 2000;32:47–50. 15. Matsuda M, Shimomura I, Sata M, Arita Y, Nishida M, Maeda N, Okamoto M, Nagaretani H, Nishizawa H, Kishida K, Komuro R, Kihara S, Nagai R, Funahashi T, Matsuzawa Y: Role of adiponectin in preventing vascular stenosis. The missing link of adipo-vascular axis. J Biol Chem 2002;277:37487–37491. 16. Pischon T, Girman CJ, Hotamisligil GS, Rifai N, Hu FB, Rimm EB: Plasma adiponectin levels and risk of myocardial infarction in men. JAMA 2004;291:1730 –1777. 17. Zoccali C, Mullamaci F, Tripepi G, Benedetto FA, Cuturupi S, Parlongo S: Adiponectin, metabolic risk facotrs and cardiovascular events among patient with end-stage renal disease. Am J Soc Nephrol 2002;13:134 –141. 18. Tentolouris N, Doulgerakis D, Moyssakis I, Kyriaki D, Makrilakis K, Kosmadakis G, Stamatiadis D, Katsilambros N, Stathakis C: Plasma adiponectin concentrations in patients with chronic renal failure: relationship with metabolic risk factors and ischaemic heart disease. Horm Metab Res 2004;36:721–727. 19. Schulze MB, Shai I, Rimm EB, Li T, Rifai N, Hu FB: Adiponectin and future coronary heart disease events among men with type 2 diabetes. Diabetes 2005;54:534 –539. 20. Adamczak M, Wiecek A, Funahashi T, Chudek J, Kokot F, Matsuzawa Y: Decreased plasma adiponectin concentration in patients with essential hypertension. Am J Hypertens 2003;16:72–75. 21. Mallamaci F, Zoccali C, Cuzzola F, Tripepi G, Cutrupi S, Parlongo S, Tanaka S, Ouchi N, Kihara S, Funahashi T, Matsuzawa Y: Adiponectin in essential hypertension. J Nephrol 2002;15:507–511. 22. Furuhashi M, Ura N, Higashiura K, Murakami H, Tanaka M,
AJH–December 2005–VOL. 18, NO. 12
23.
24.
25. 26. 27.
28.
29. 30.
31.
Moniwa N, Yoshida D, Shimamoto K: Blockade of the renin-angiotensin system increases adiponectin concentrations in patients with essential hypertension. Hypertension 2003;42:76 – 81. Mea PD, Lupia M, Bandolin V, Guzzon S, Sonino N, Vettor R, Fallo F: Adiponectin, insulin resistance and left ventricular structure in dipper and non-dipper essential hypertensive patients. Am J Hypertens 2005;18:30 –35. Yasmin ??, McEniery CM, Wallace S, Mackenzie IS, Cockcroft JR, Wilkinson IB: C-reactive protein is associated with arterial stiffness in apparently healthy individuals. Arterioscler Thromb Vasc Biol 2004;24:969 –974. Hug C, Lodish HF: The role of the adipocyte hormone adiponectin in cardiovascular disease. Curr Opin Pharmacol 2005;5:1– 6. Mahmud A, Feely J: Effect of smoking on arterial stiffness and pulse pressure amplification. Hypertension 2003;41:183–187. Kazumi T, Kawaguchi A, Sakai K, Hirano T, Yoshino G: Young men with high–normal blood pressure have lower serum adiponectin, smaller LDL size and higher elevated heart rate than those with optimal blood pressure. Diabetes Care 2002;25:971–976. Festa A, D’Agostini R, Hales CN, Mykkanen L, Haffner SM: Heart rate in relation to insulin sensitivity and insulin secretion in nondiabetic subjects. Diabetes Care 2000;23:624 – 628. Julius S, Valentini M, Palatini P: Overweight and hypertension: a two-way street. Hypertension 2000;35:807– 813. Wakabayashi S, Aso Y: Adiponectin concentrations in sera from patients with type 2 diabetes are negatively associated with sympathovagal balance as evaluated by power spectral analysis of heart rate variation. Diabetes Care 2004;27:2392–2397. Mahmud A, Feely J: Spurious systolic hypertension of youth: fit young men with elastic arteries. Am J Hypertens 2003;16:229 –232.
2005; 18:1543–1548
Blood Vessels
Adiponectin and Arterial Stiffness Azra Mahmud and John Feely Background: Adiponectin, an anti-inflammatory adipocytokine, is reduced in hypertension, diabetes, and coronary artery disease (CAD). Arterial stiffness, as aortic pulse wave velocity (PWV) in hypertension and diabetes, and as augmentation index (AIx) in CAD, is independently associated with cardiovascular mortality. We explored the relationship between adiponectin and arterial stiffness in essential hypertension. Methods: Seventy-six untreated patients, 34 women, aged 47 ⫾ 1 years, mean ⫾ SEM with essential hypertension, had blood pressure (BP), carotid–femoral PWV, AIx plasma adiponectin, and proinflammatory cytokine C-reactive protein (CRP) measured using ELISA technique after an overnight fast. Results were analyzed using univariate and multiple logistic regression analysis. Results: There was a significant positive relationship between log adiponectin and AIx (r ⫽ 0.33, P ⬍ .005) and plasma HDL-cholesterol (r ⫽ 0.40, P ⬍ .001). In contrast there were significant negative relationships with PWV (r
⫽ ⫺0.24, P ⬍ .05), transit time (r ⫽ ⫺0.37, P ⬍ .001), and pulse pressure amplification (r ⫽ ⫺0.30, P ⬍ .005) in addition to waist circumference, waist-to-hip ratio, height, and weight. In a stepwise regression model, the independent predictors of AIx were heart rate, height, mean arterial pressure, age, and gender (R2 ⫽ 0.69, P ⬍ .0001) with no contribution from adiponectin. However, for PWV (R2 ⫽ 0.59, P ⬍ .0001) the independent determinants were mean arterial pressure, age, and adiponectin. Conclusions: These results show a divergent relationship between adiponectin and arterial stiffness, negative for PWV, and positive for wave reflection (AIx). Anthropomorphic factors, particularly height, weight, and heart rate may influence the relationship to the latter. Adiponectin is an independent predictor of aortic PWV but not of AIx. Am J Hypertens 2005;18:1543–1548 © 2005 American Journal of Hypertension, Ltd. Key Words: Adiponectin, arterial stiffness, hypertension.
diponectin has emerged in the past few years as an important adipocytokine modulating insulin resistance with anti-inflammatory and antiatherogenic properties.1 It is expressed and excreted only in the adipose tissue and plasma levels are relatively high, representing approximately 0.01% of total plasma proteins and concentrations exceed those of many other common hormones. Adiponectin levels are related positively to HDLcholesterol but inversely to insulin1 and its levels are reduced in obesity and type 2 diabetes.2 Insulin has inhibitory effects on the levels of adiponectin.1 More recently, adiponectin has been shown to possess anti-inflammatory properties and may modulate the process of atherogenesis,1 suggesting that adiponectin deficiency associated with obesity, metabolic syndrome, and type 2 diabetes may contribute to accelerated atherosclerosis in such conditions. Atherosclerotic changes in large arteries make an important contribution to pathogenesis of cardiovascular disease. Increased aortic stiffness measured as carotid–
A
femoral pulse wave velocity (PWV) is associated with increased risk of cardiovascular events in the hypertensive3 and diabetic4 populations. Augmentation index (AIx), a composite of wave reflection from medium-sized muscular arteries, left ventricular ejection, and PWV, is related to the development of coronary artery disease (CAD).5 The PWV is significantly correlated with plasma glucose levels in subjects with impaired glucose tolerance6 and there is a significant relationship between plasma insulin levels and aortic stiffness in the general population7 and in subjects with metabolic syndrome.8 There is also a significant association between the changes in brachial ankle PWV and adiponectin in type 2 diabetes patients after treatment with pioglitazone, which significantly reduced both brachial ankle PWV and increased adiponectin concentrations.9 However, there is no information on the relationship between AIx and adiponectin in hypertension. Therefore, the aim of our study was to explore the relationship between adiponectin and arterial stiffness determined by both PWV, a classic marker of aortic stiffness
Received March 14, 2005. First decision June 9, 2005. Accepted June 10, 2005. From the Department of Therapeutics and Hypertension Clinic, Trinity Centre for Health Sciences, St. James’s Hospital, Dublin, Ireland.
Address correspondence and reprint requests to Professor John Feely, Department of Therapeutics and Hypertension Clinic, Trinity Centre for Health Sciences, St. James’s Hospital, Dublin 8, Ireland; e-mail: [email protected]
© 2005 by the American Journal of Hypertension, Ltd. Published by Elsevier Inc.
0895-7061/05/$30.00 doi:10.1016/j.amjhyper.2005.06.014
1544
ADIPONECTIN AND ARTERIAL STIFFNESS
and AIx, reflecting the stiffness of the medium-sized muscular arteries, in a never-treated hypertensive population.
Methods Patients and Methods We recruited 76 never-treated hypertensive subjects with diagnosis based on three outpatient measures of blood pressure (BP) greater than 140/90 mm Hg and confirmed by daytime ambulatory BP monitoring (⬎135/80 mm Hg). The mean (⫾SEM) age was 47 ⫾ 1 years (range 38 to 64 years), 34 were women and 26% of the patients were smokers. None of these subjects had received treatment for hypertension or any other medications that may influence BP or arterial stiffness. None of the patients had secondary causes of hypertension, CAD, renal disease, liver disorders, or diabetes. Patients were studied in the morning after an overnight fast having refrained from caffeinecontaining beverages, alcohol, and smoking in the previous 12 h. The subjects gave informed consent and the study had institutional ethics committee approval. Laboratory Measurements Height, weight, waist, and hip measurements were recorded for each patient. Venous samples were drawn into EDTA tubes, centrifuged (4°C, 2500 rpm) for 10 min and supernatant stored at ⫺80°C. Plasma adiponectin, C-reactive protein (CRP), tumor necrosis factor-␣ (TNF-␣), and interleukin-6 (IL-6) levels were measured by enzymelinked immunosorbent assay (ELISA) (R & D Systems Ltd., Abingdon, Oxford, UK) in duplicates. In addition, the plasma levels of total cholesterol, LDL, HDL, glucose, and creatinine were measured by standard methods and the creatinine clearance calculated. Hemodynamic Measurements After a supine rest of 15 min, BP and heart rate was measured with an automated oscillometric device (Omron model HEM 705 CP, Omron Corporation, Tokyo, Japan) in the right arm after which AIx and carotid femoral PWV were measured as previously described.3 Arterial Pulse Wave Analysis Radial applanation tonometry was used to acquire the aortic pressure waveform noninvasively using the generalized transfer fraction (Sphygmocor, version 7.1, Sydney, Australia). The aortic pressure waveform was used to derive the aortic BP and the AIx (difference in height of the first and second systolic peak in the aorta and aortic pulse pressure). Time to reflected wave (TR) was measured from the foot of the wave to the inflection point on the waveform and left ventricular ejection was measured from the foot of the wave to the diastolic incisura. The pulse pressure amplification was calculated as the difference between brachial and aortic pulse pressure.
AJH–December 2005–VOL. 18, NO. 12
Pulse Wave Velocity Carotid–femoral PWV was measured with an automated system (Artech Medical, Coléon, France) using the footto-foot method. The carotid and femoral waveforms were acquired simultaneously with two pressure-sensitive transducers and the transit time of the pulse calculated by the system software. The distance between the two arterial sites was measured on the body using a tape measure and PWV calculated as the distance divided by time (meters per seconds). At least 12 successive readings were used for analysis to cover a complete respiratory cycle. Statistical Analysis Statistical analysis performed with JMP version 5.0 (SAS for Windows, Cary, NC, USA). As PWV adiponectin and the inflammatory cytokines were not normally distributed they were log transformed for further analysis. The coefficient of variation of all the primary measures PWV, AIx, adiponectin, and inflammatory markers was under 5%. Relationship between adiponectin, PWV, AIx, and other parameters was analyzed using nonparametric methods (Spearman correlations). To study whether there was an independent relationship between PWV and AIx with adiponectin, stepwise regression analysis was used. In the first model, we took PWV as the dependent variable and age, mean arterial pressure, gender, body mass index (BMI), waist/hip ratio, heart rate, smoking status, adiponectin and CRP, plasma glucose, and lipoproteins as independent variables. In the second model, with the same independent variables, AIx was analyzed as the dependent variable. Results are expressed as mean ⫾ SEM; P ⬍ .05 considered significant.
Results Baseline data for the patient population is summarized in Tables 1 and 2. The relationship between adiponectin and measures of arterial stiffness and other hemodynamic variables and anthropological parameters is shown in Table 3. Adiponectin was inversely related to subject’s height, weight, waist, waist-to-hip ratio, but not BMI. There was also an inverse correlation between PWV, TR, pulse pressure amplification, and adiponectin. In contrast the significant relationship with AIx was positive. There was no relationship between adiponectin and brachial or aortic BP, with the brachial diastolic BP being closest to achieving statistical significance (P ⫽ .07). Although the relationship with heart rate was not statistically significant there was a positive relationship with left ventricular ejection duration. Adiponectin was positively correlated with HDL-cholesterol and negatively with glucose and triglycerides, but not with total cholesterol. There was no significant relationship between adiponectin levels and CRP (r ⫽ ⫺0.08), TNF-␣ (r ⫽ ⫺0.10), and IL-6 (r ⫽ ⫺0.13). The relationship between PWV and other parameters is given in Table
AJH–December 2005–VOL. 18, NO. 12
ADIPONECTIN AND ARTERIAL STIFFNESS
Table 1. Baseline characteristics of the patient population (n ⫽ 76) Parameter Age (yr) Gender (M/F) Height (cm) Weight (kg) Body mass index (kg/m2) Waist (cm) Hip (cm) Waist/hip ratio Brachial systolic BP (mm Hg) Brachial diastolic BP (mm Hg) Mean arterial pressure (mm Hg) Brachial pulse pressure (mm Hg) Heart rate (min⫺1) Aortic systolic BP (mm Hg) Aortic diastolic BP (mm Hg) Aortic pulse pressure (mm Hg) Total cholesterol (mmol/L) HDL-cholesterol (mmol/L) LDL-cholesterol (mmol/L) Triglycerides (mmol/L) Glucose (mmol/L) Serum creatinine (mmol/L) Creatinine clearance (mL/min)
Value (ⴞ mean SEM) 47 ⫾ 1 (44/32) 171 ⫾ 0.1 88.6 ⫾ 2 30 ⫾ 0.5 96.5 ⫾ 2 105 ⫾ 1 0.92 ⫾ 0.01 152 ⫾ 2 92 ⫾ 1 114 ⫾ 2 62 ⫾ 2 70 ⫾ 1 141 ⫾ 2 93 ⫾ 1 48 ⫾ 1.6 5.17 ⫾ 0.1 1.27 ⫾ 0.03 2.95 ⫾ 0.1 1.76 ⫾ 0.1 5.6 ⫾ 0.1 88.6 ⫾ 2 109 ⫾ 4
1545
Table 3. Univariate relationship between plasma adiponectin concentrations and hemodynamic, biochemical, and anthropomorphic variables (n ⫽ 76) Variables Pulse wave velocity (m/sec) Augmentation index (%) TR (msec) Pulse pressure amplification (mm Hg) Left ventricular ejection duration (msec) Weight (kg) Height (cm) Waist (cm) Waist/hip ratio Plasma glucose (mmol/L) Plasma triglycerides (mmol/L) Plasma HDL (mmol/L)
Correlation Coefficient
P
⫺0.24 0.33 ⫺0.37
⬍.05 ⬍.01 ⬍.001
⫺0.30
⬍.01
0.28 ⫺0.32 ⫺0.55 ⫺0.23 ⫺0.42
⬍.01 ⬍.01 ⬍.001 ⬍.05 ⬍.001
⫺0.23
⬍.05
⫺0.25 0.40
⬍.05 ⬍.0001
arterial pressure, age, and gender (R2 ⫽ 0.69, P ⬍ .0001) with no contribution from adiponectin.
Discussion 4. There was no difference in plasma adiponectin levels, PWV, and AIx between smokers and nonsmokers. The PWV was significantly and positively related to age, BP, AIx, and plasma glucose and inversely to TR and HDLcholesterol with no relationship with heart rate, BMI, waist, and hip measurements or LDL and total cholesterol. In a stepwise multiple regression model, plasma adiponectin levels independently predicted PWV, when adjusted for age, mean arterial pressure, gender, height, weight, waist-to-hip ratio, conventional cardiovascular risk factors, and CRP with age and mean arterial pressure the other independent predictors (R2 ⫽ 0.62, P ⬍ .0001; Table 5). However, when adjusted for the subject’s anthropomorphic measurements, cardiovascular risk factors and CRP, AIx was predicted by heart rate, height, mean
Our results show a significant inverse relationship between plasma adiponectin levels and PWV in hypertensive subjects. The finding of a negative relationship with PWV is in keeping with previous observations relating both PWV and adiponectin to cardiovascular risk. An increased PWV is associated with cardiovascular events in the hypertensive3 and diabetic4 populations and in subjects with endstage renal disease (ESRD).10 Low plasma adiponectin concentrations may predict the risk of acute coronary syndrome,11 but not of restenosis after coronary stenting.12 Adiponectin modulates endothelial adhesion molecules13 and has been found in the subendothelial space of the subcarotid arteries injured by a catheter and in atheroscle-
Table 4. Univariate correlations for pulse wave velocity Table 2. Baseline parameters for arterial stiffness, adiponectin, and markers of inflammation C-reactive protein, tumor necrosis factor (TNF)-␣, and interleukin (IL-6) Pulse wave velocity (m/sec) Augmentation index (%) Transit time (msec) Pulse pressure amplification (mm Hg) Adiponectin (g/mL) C-reactive protein (mg/L) TNF-␣ (pg/mL) IL-6 (pg/mL)
10.4 ⫾ 0.2 27 ⫾ 1.3 137 ⫾ 1.6 12.5 8.2 0.36 1.0 1.7
⫾ ⫾ ⫾ ⫾ ⫾
0.7 0.4 0.05 0.04 0.2
Variables Age Brachial systolic BP (mm Hg) Brachial diastolic BP (mm Hg) Mean arterial pressure (mm Hg) Augmentation index (%) Transit time (msec) Plasma glucose (mmol/L) Plasma HDL (mmol/L)
Correlation Coefficient
P
0.56
⬍.0001
0.74
⬍.0001
0.50
⬍.001
0.65 0.26 ⫺0.26 0.30 ⫺0.26
⬍.0001 ⬍.05 ⬍.05 ⬍.05 ⬍.05
1546
ADIPONECTIN AND ARTERIAL STIFFNESS
AJH–December 2005–VOL. 18, NO. 12
Table 5. Multiple regression models with pulse wave velocity (PWV R2 ⫽ 0.59, P ⬍ .001) and augmentation index (Alx R2 ⫽ 0.69, P ⬍ .001) as the dependent variable Parameters for PWV Mean arterial pressure (mm Hg) Age (yr) Adiponectin (g/mL) Parameters for AIx Heart rate (min⫺1) Height (cm) Mean arterial pressure (mm Hg) Age (yr) Gender (female)
R2

SE
P
0.41 0.53 0.57
0.06 0.07 ⫺0.10
0.01 0.01 0.04
⬍.0001 ⬍.0001 ⬍.01
0.21 0.42 0.62 0.66 0.71
⫺0.51 ⫺0.28 0.33 0.29 3.0
0.06 0.11 0.06 0.07 1.1
⬍.0001 ⬍.01 ⬍.0001 ⬍.0001 ⬍.01
rotic lesions within the endothelium14 and may prevent vascular restenosis.15 Circulating adiponectin concentrations are protective against CAD16 and also predictive of subsequent cardiovascular events in patient with ESRD.17 Aortic distensibility measured by ultrasound was lower in patients with chronic renal failure, suggestive of increased stiffness, but there was no relationship between distensibility and adiponectin in either controls or the patients with renal disease.18 However, in diabetic patients where adiponectin levels are reduced2 PWV is increased. Reduced adiponectin levels are associated with increased future coronary heart disease events in men with type 2 diabetes.19 In subjects with type 2 diabetes there is a significant correlation between changes in adiponectin concentrations and changes in the ankle brachial pulse wave estimate of PWV9 after treatment with pioglitazone. Studies looking at adiponectin concentrations in patients with hypertension have produced conflicting results. Adamczak et al20 reported lower concentrations, whereas Mallamaci et al21 showed higher adiponectin concentrations in hypertensive patients. Furuhashi et al22 found reduced adiponectin concentration in a group (40%) of hypertensive patients who had insulin resistance determined by using a euglycemic hyperinsulinemic glucose clamp technique. It has recently been shown23 that essential hypertensive patients with a nondipping pattern show more prominent insulin resistance and lower adiponectin levels compared to dippers. No correlation was found between adiponectin and systolic/diastolic (clinic or 24-h ambulatory) BP levels,17,21–23 although one study20 did show a negative correlation with systolic, diastolic, and mean arterial pressure. Whether the relationship between adiponectin and PWV extends to a normotensive population or a “threshold” level of pressure is required to lower adiponectin levels was not studied. Our data did not provide strong evidence of an association between adiponectin concentration and brachial or aortic BPs. This study also provides an interesting dichotomy in the relationship between indices of arterial stiffness, both independent cardiovascular risk factors, and the concentration of the adipocytokine adiponectin, which itself may play a significant role in reducing cardiovascular risk. In
contrast to PWV, the relationship between arterial stiffness as assessed by AIx or wave reflection, and adiponectin is a positive one (Figure 1). How adiponectin and PWV are related is not immediately clear and may possibly be attributable to other metabolic and vasoactive factors not measured in the study. For example, adipocytes also secrete TNF, leptin, resistin, and a number of other growth factors.1 Because there is evidence that systemic inflammation, as represented by C-reactive protein concentration, is related positively to PWV,24 one may speculate that the relationship is due to a reduced anti-inflammatory effect as a consequence of lower concentrations of adiponectin. However, we did not detect any relationship between adiponectin concentration or other markers of inflammation—TNF, IL-6, and CRP. Furthermore, adiponectin independently predicted PWV, even when CRP is added in the model. Unfortunately we have not measured insulin resistance, which is common in hypertension, but there was a negative relationship between adiponectin and plasma glucose suggesting insulin resistance, which we know increases PWV, and this may be one mechanism. Adiponectin also binds to collagen but to separate the anti-inflammatory and insulin resistance components of adiponectin it may be necessary to measure subtypes, trimers, and the carboxy-terminal globular domain that activates AMP-activated protein kinase and the hexamer and high molecular isoforms that activate nuclear factor- signaling pathways.25 We did not observe any difference between smokers and nonsmokers in relation to plasma adiponectin levels, PWV, or AIx in this study, although we have reported previously a significantly higher AIx in a young healthy normotensive smoking cohort compared to the nonsmoking group.26 However, this may be due to the small number in our study of middle-aged hypertenisves and as suggested earlier,26 an age-related effect of smoking on BP. Also the effect of smoking on arterial stiffening may be quite different in healthy compliant arteries compared to hypertensive patients with established arterial stiffening. An examination of the determinants of the two parameters of stiffness that we measured may provide some insight as to their divergent relationship with adiponectin. As previously reported3 and seen in this study age and BP
AJH–December 2005–VOL. 18, NO. 12
ADIPONECTIN AND ARTERIAL STIFFNESS
1547
uals have higher adiponectin and higher AIx, the latter in part due to the shorter distance for wave travel to reflection sites. It is well established31 that taller individuals have not only lower AIx but also increased pulse pressure amplification. Therefore, the negative relationship to amplification in this study suggests that these hemodynamic influences may be determinants for the positive relationship between AIx and adiponectin. In multivariate analysis adiponectin is not an independent determinant of AIx (in contrast to PWV) and its relationship is one of association with the other, particularly with anthropomorphic factors determining AIx. This study also emphasizes the need to have complimentary but independent measures of arterial stiffness.
References 1. 2.
3.
4.
5. FIG. 1. Relationship between plasma adiponectin and pulse wave velocity (top) and augmentation index (bottom) in patients with essential hypertension (n ⫽ 76).
are the major determinants of PWV. Both age and BP also contribute to AIx, which is a composite of the extent and site of wave reflection, the rate of wave propagation— largely PWV and ventricular ejection. However, arterial wave reflection is influenced particularly by anthropomorphic (eg, height, gender) factors in addition to hemodynamic factors including heart rate and site of the reflected wave. In this study heart rate and height were important independent determinants of AIx (Table 5), and although the negative correlation between adiponectin and heart rate was not significant, it was positively correlated with ejection duration. Healthy young men (college students) with high normal BP have lower serum adiponectin and higher heart rates than men with optimal BP, even after adjustment for BMI.27 Heart rate is inversely and independently related to insulin sensitivity.28 Sympathetic overactivity may be a common preceding feature of higher heart rate and insulin resistance.27,29 Relative sympathetic activation, as derived from heart rate variation, was associated with low adiponectin in patients with non-insulin-dependent diabetes mellitus.30. It is clear from the anthropomorphic measurements that the strongest correlation with adiponectin concentrations is body height. Shorter individ-
6.
7.
8.
9.
10.
11.
12.
13.
Gil-Campos M, Canete R, Gil A: Adiponectin, the missing link in insulin resistance and obesity. Clin Nutr 2004;23:963–974. Hotta K, Funahashi T, Arita Y: Plasma concentrations of a novel, adipose-specific protein, adiponectin, in type 2 diabetic patients. Arterioscl Thromb Vasc Biol 2000;20:1595–1599. Safar ME, Henry O, Meaume S: Aortic pulse wave velocity: an independent marker of cardiovascular risk. Am J Ger Cardiol 2002; 11:295–298. Cruickshank K, Riste L, Anderson SG, Wright JS, Dunn G, Gosling RG: Aortic pulse wave velocity and its relationship to mortality in diabetes and glucose intolerance: an integrated index of vascular function? Circulation 2002;106:2085–2090. Weber T, Auer J, O’Rourke MF, Kvas E, Lassnig E, Berent R, Eber B: Arterial stiffness, wave reflections and the risk of coronary artery disease. Circulation 2004;109:184 –189. Ohnishi H, Saitoh S, Takagi S: Pulse wave velocity as an indicator of atherosclerosis in impaired fasting glucose: the Tanno and Sobetsu study. Diabetic Care 2003;26:437– 440. Hansen TW, Jeppessen J, Rasmussen S, Ibsen H, Torp-Pedersen C: Relation between insulin and aortic stiffness: a population based study. J Hum Hypertens 2004;18:1–7. Choi KM, Lee KW, Soe JA, Oh JH, Kim SG, Kim NH, Choi DS, Baik SH: Relationship between brachial ankle pulse wave velocity and cardiovascular risk factors of the metabolic syndrome. Diabetes Res Clin Prac 2004;66:57– 61. Satoh N, Ogawa Y, Usui T, Tagami T, Kono S, Uesugi H, Sugiyama H, Sugawara A, Yamada K, Shimatsu A, Kuzuya H, Nakao K: Antiatherogenic effect of pioglitazone in type 2 diabetic patients irrespective of the responsiveness to its antidiabetic effect. Diabetes Care 2003;26:2493–2499. Blacher J, London GM, Safar B, Mourad J-J: Influence of age and end-stage renal disease on the stiffness of carotid wall material in hypertension. J Hypertens 1999;17:237–244. Nakamura Y, Shimada K, Fukuda D, Shimada Y, Ehara S, Hirose M, Kataoka T, Kamimori K, Shimodozono S, Kobayashi Y, Yoshiyama M, Takeuchi K, Yoshikawa J: Implication of plasma concentrations of adiponectin in patients with coronary artery disease. Heart 2004:90:528 –533. Shimada K, Miyauchi K, Mokuno H, Miyazaki T, Seki E, Watanabe Y, Iwama Y, Shigekiyo M, Matsumoto M, Okazaki S, Tanimoto K, Kawamura M, Suzuki H, Kurata T, Sato H, Daida H: Predictive value of the adipocyte-derived plasma protein adiponectin for restenosis after elective coronary stenting. Jpn Heart J 2002;43:85–91. Ouchi N, Kihara S, Arita Y: Novel modulator for endothelial adhesion molecules: adipocyte-derived plasma protein adiponectin. Circulation 1999;100:2473–2476.
1548
ADIPONECTIN AND ARTERIAL STIFFNESS
14. Okamoto Y, Arita Y, Nishida M: An adipocyte-derived plasma protein, adiponectin, adheres to injured vascular walls. Horm Metab Res 2000;32:47–50. 15. Matsuda M, Shimomura I, Sata M, Arita Y, Nishida M, Maeda N, Okamoto M, Nagaretani H, Nishizawa H, Kishida K, Komuro R, Kihara S, Nagai R, Funahashi T, Matsuzawa Y: Role of adiponectin in preventing vascular stenosis. The missing link of adipo-vascular axis. J Biol Chem 2002;277:37487–37491. 16. Pischon T, Girman CJ, Hotamisligil GS, Rifai N, Hu FB, Rimm EB: Plasma adiponectin levels and risk of myocardial infarction in men. JAMA 2004;291:1730 –1777. 17. Zoccali C, Mullamaci F, Tripepi G, Benedetto FA, Cuturupi S, Parlongo S: Adiponectin, metabolic risk facotrs and cardiovascular events among patient with end-stage renal disease. Am J Soc Nephrol 2002;13:134 –141. 18. Tentolouris N, Doulgerakis D, Moyssakis I, Kyriaki D, Makrilakis K, Kosmadakis G, Stamatiadis D, Katsilambros N, Stathakis C: Plasma adiponectin concentrations in patients with chronic renal failure: relationship with metabolic risk factors and ischaemic heart disease. Horm Metab Res 2004;36:721–727. 19. Schulze MB, Shai I, Rimm EB, Li T, Rifai N, Hu FB: Adiponectin and future coronary heart disease events among men with type 2 diabetes. Diabetes 2005;54:534 –539. 20. Adamczak M, Wiecek A, Funahashi T, Chudek J, Kokot F, Matsuzawa Y: Decreased plasma adiponectin concentration in patients with essential hypertension. Am J Hypertens 2003;16:72–75. 21. Mallamaci F, Zoccali C, Cuzzola F, Tripepi G, Cutrupi S, Parlongo S, Tanaka S, Ouchi N, Kihara S, Funahashi T, Matsuzawa Y: Adiponectin in essential hypertension. J Nephrol 2002;15:507–511. 22. Furuhashi M, Ura N, Higashiura K, Murakami H, Tanaka M,
AJH–December 2005–VOL. 18, NO. 12
23.
24.
25. 26. 27.
28.
29. 30.
31.
Moniwa N, Yoshida D, Shimamoto K: Blockade of the renin-angiotensin system increases adiponectin concentrations in patients with essential hypertension. Hypertension 2003;42:76 – 81. Mea PD, Lupia M, Bandolin V, Guzzon S, Sonino N, Vettor R, Fallo F: Adiponectin, insulin resistance and left ventricular structure in dipper and non-dipper essential hypertensive patients. Am J Hypertens 2005;18:30 –35. Yasmin ??, McEniery CM, Wallace S, Mackenzie IS, Cockcroft JR, Wilkinson IB: C-reactive protein is associated with arterial stiffness in apparently healthy individuals. Arterioscler Thromb Vasc Biol 2004;24:969 –974. Hug C, Lodish HF: The role of the adipocyte hormone adiponectin in cardiovascular disease. Curr Opin Pharmacol 2005;5:1– 6. Mahmud A, Feely J: Effect of smoking on arterial stiffness and pulse pressure amplification. Hypertension 2003;41:183–187. Kazumi T, Kawaguchi A, Sakai K, Hirano T, Yoshino G: Young men with high–normal blood pressure have lower serum adiponectin, smaller LDL size and higher elevated heart rate than those with optimal blood pressure. Diabetes Care 2002;25:971–976. Festa A, D’Agostini R, Hales CN, Mykkanen L, Haffner SM: Heart rate in relation to insulin sensitivity and insulin secretion in nondiabetic subjects. Diabetes Care 2000;23:624 – 628. Julius S, Valentini M, Palatini P: Overweight and hypertension: a two-way street. Hypertension 2000;35:807– 813. Wakabayashi S, Aso Y: Adiponectin concentrations in sera from patients with type 2 diabetes are negatively associated with sympathovagal balance as evaluated by power spectral analysis of heart rate variation. Diabetes Care 2004;27:2392–2397. Mahmud A, Feely J: Spurious systolic hypertension of youth: fit young men with elastic arteries. Am J Hypertens 2003;16:229 –232.