ARTICLE IN PRESS

Nutrition, Metabolism & Cardiovascular Diseases (2007) xx, 1e8

+ MODEL

ava i lab le at www.sc ienced i rec t . com

journa l homepage : www.e l sev i er . com/ loca te /nmcd

Endothelial function and arterial stiffness innormotensive normoglycemic first-degreerelatives of diabetic patients are independentof the metabolic syndrome

Angelo Scuteri b,*, Manfredi Tesauro a, Stefano Rizza a, Micaela Iantorno a,Massimo Federici a, Davide Lauro a, Umberto Campia a, Mario Turriziani a,Angelo Fusco a, Giulio Cocciolillo a, Renato Lauro a

a Internal Medicine Department, University of Tor Vergata, Rome, Italyb UO Geriatria, INRCAeIRCCS, Via Cassia 1167, 00189 Rome, Italy

Received 3 October 2006; received in revised form 31 January 2007; accepted 12 March 2007

KEYWORDSEndothelium;Arteries;Aging;Diabetes mellitus;Metabolic syndrome

* Corresponding author. Fax: þ39 06E-mail address: angeloelefante@in

0939-4753/$ - see front matter ª 200doi:10.1016/j.numecd.2007.03.008

Please cite this article in press as: Screlatives of diabetic patients arej.numecd.2007.03.008

Abstract Background and aim: The aim of the present study was to investigate endothelialfunction and arterial stiffness in normotensive normoglycemic first-degree relatives (offspring)of diabetic subjects and to explore the relationship with the metabolic syndrome and itscomponents.Methods and results: Forty-five healthy normotensive normoglycemic subjects (aged 18e

42 years), 29 first-degree relatives of diabetic subjects (FDR) and 16 with no parental historyof type 2 diabetes mellitus were studied. Endothelial function was measured as flow-mediateddilation of the brachial artery (FMD) and arterial stiffness as carotid-femoral pulse wave veloc-ity (PWV). Insulin resistance was calculated by homeostasis model assessment (HOMA). Plasmalevels of inflammation markers (hsCRP, TNF-a, IL-1b, CD40L, VCAM, and ICAM) were evaluated.

Normotensive normoglycemic FDR presented a 33% lower flow-mediated dilation than thecontrol group (9.8 � 5.2 vs. 16.2 � 7.6%, p < 0.01). FMD was reduced in FDR, with or withoutinsulin resistance, whereas arterial stiffness was significantly increased only in FDR with insulinresistance. To investigate the role of FDR status independently of altered components of themetabolic syndrome, subjects with no altered components of the metabolic syndrome werecompared according to their FDR status: FDR subjects with no altered components of themetabolic syndrome presented a blunted endothelial function (lower FMD: 11.2 � 1.6 vs.16.8 � 2.0%, p < 0.05) and stiffer large arteries (higher PWV: 9.6 � 0.3 vs. 8.8 � 0.3 m/s,p < 0.05) than controls.

3036 2896.terfree.it (A. Scuteri).

7 Elsevier B.V. All rights reserved.

uteri A et al., Endothelial function and arterial stiffness in normotensive normoglycemic first-degreeindependent of the metabolic syndrome, Nutr Metab Cardiovasc Dis (2007), doi:10.1016/

ARTICLE IN PRESS

2 A. Scuteri et al.

+ MODEL

Please cite this article in press as: Screlatives of diabetic patients arej.numecd.2007.03.008

Conclusion: Normoglycemic first-degree relatives of diabetic subjects have blunted endothe-lial function and increased stiffness of the large arteries. These alterations are already presentat a very young age, before any alteration in glycemic control or blood pressure values can bedetected, and are independent of the presence of the metabolic syndrome and its alteredcomponents.ª 2007 Elsevier B.V. All rights reserved.

Introduction

Diabetes mellitus is a major risk factor for cardiovascularmorbidity and mortality [1,2]. The two- to four-fold in-creased risk of cardiovascular disease among diabetic sub-jects has been attributed to more severe vascular damagein diabetic subjects [3], often occurring before hyperglyce-mia becomes evident [4].

Endothelial dysfunction [5,6] and arterial stiffening[7e9] have been suggested as possible mechanisms ofincreased risk of cardiovascular events. Endothelial dys-function, detected non-invasively as a decreased flow-mediated dilation (FMD) of the brachial artery [10,11], isan early and potentially reversible functional disturbancein the development of vascular lesion and disease [12]. Itis also common in subjects with the metabolic syndromeand it is believed to reflect the metabolic derangementspresent in this condition [13]. We have recently reportedthat the metabolic syndrome is associated with increasedthickness and stiffness of the carotid artery, independentlyof each single component of this syndrome [14].

Whether family history for diabetes predispose toalteration in vascular function through an unknown ‘‘ge-netic susceptibility’’ or through higher prevalence ofaltered components of the metabolic syndrome has notyet been investigated.

The aim of the present study was to investigateendothelial function and arterial stiffness in normotensivenormoglycemic first-degree relatives (offspring) of diabeticsubjects and to explore their relationship with the meta-bolic syndrome and its components.

Methods

The study population consisted of 45 subjects (aged 18e42 years): 29 healthy normotensive normoglycemic first-degree relatives of diabetic subjects (one parent withtype 2 diabetes mellitus) (17 men and 12 women) and 16healthy normotensive normoglycemic subjects with noparental history of type 2 diabetes mellitus (7 men and 9women). Participants were randomly recruited via theirparents from our outpatient clinic. Exclusion criteriawere clinically manifest metabolic, cardiovascular or anyother serious chronic disease. Written informed consentwas obtained from all subjects. All women were pre-menopausal and their investigations were undertakenduring the first week of their menstrual cycles. None ofthem were taking oral contraceptives.

Normotension was defined as SBP <140 and/or DBP<90 mmHg in the absence of use of concomitant

uteri A et al., Endothelial functionindependent of the metabolic

antihypertensive medications [15]. Normoglycemic statuswas defined as a fasting glucose <110 mg/dL in the absenceof use of concomitant antidiabetic medications [16].

Participants were asked to refrain from drinking alcohol,smoking or beverages containing caffeine for at least 24 hbefore the study.

Anthropometric, metabolic characteristicsand laboratory measurements

Height, weight and waist circumference were determinedfor all participants. Body mass index (BMI) was determinedas body weight (kg)/height (m)2. Total and HDL cholesterol,triglycerides, and routine chemistry were measured bystandard laboratory techniques. Plasma glucose was mea-sured by the enzymatic in vitro test (Roche, automatedchemistry clinical analyzer). Serum insulin was determinedby an electrochemiluminescense assay (Eclia, Roche). Insu-lin resistance was calculated by homeostasis model assess-ment (HOMA) [17]. An oral glucose tolerance test (OGTT)was performed in each participant according to the recom-mendations of the American Diabetes Association (ADA)Expert Committee on the Classification and Diagnosis ofDiabetes Mellitus [17].

Blood pressure measurement

Supine blood pressure was measured using a manual sphyg-momanometer. After a 10-min rest period, blood pressurewas measured 3 times, and the average of the last 2measurements was used for statistical analyses. Brachialpulse pressure (PP) was calculated as the differencebetween systolic and diastolic BP. Mean arterial pressure(MBP) was calculated as DBP þ 1/3 PP.

Definition of the metabolic syndrome

The Third Report of the National Cholesterol EducationProgram Expert Panel on Detection, Evaluation, andTreatment of High Blood Cholesterol in Adults (ATP III)[18] defined the metabolic syndrome as an alteration inthree or more of the following five components: abdominalobesity (W), triglycerides (T), HDL cholesterol (H), bloodpressure (systolic or diastolic) (B), and fasting glucose(G). According to this definition, the following cut-offvalues to define alterations are used: waist circumference>102 cm for men or >88 cm for women, triglycerides�150 mg/dL, HDL cholesterol <40 mg/dL for men or<50 mg/dL for women, blood pressure �130/�85 mmHg,and fasting glucose �110 mg/dL.

and arterial stiffness in normotensive normoglycemic first-degreesyndrome, Nutr Metab Cardiovasc Dis (2007), doi:10.1016/

Table 1 Demographic profile of the study population byFDR status

Controls First-degree

n 16 29Age (years) 29 � 5 33 � 7Female sex (%) 56 41Current smoker (%) 19 27BMI (kg/m2) 22.8 � 2.2 26.8 � 5.4**Waist (cm) 80.5 � 7.7 90.9 � 13.2**SBP (mmHg) 116.5 � 13.6 118.7 � 15.8DBP (mmHg) 72.6 � 10.6 74.7 � 13.0MBP (mmHg) 87.1 � 10.7 89.2 � 13.4PP (mmHg) 43.9 � 9.7 44.0 � 8.4HR (bpm) 68.2 � 6.9 67.4 � 9.8Total cholesterol (mg/dL) 178.1 � 22.7 191.2 � 36.7LDL cholesterol (mg/dL) 114.0 � 27.1 126.0 � 36.2HDL cholesterol (mg/dL) 66.1 � 15.5 57.8 � 15.3*Triglycerides (mg/dL) 73.6 � 21.1 115.8 � 77.0*Fasting glucose (mg/dL) 86.9 � 5.9 92.1 � 9.9*2 h OGTT glucose (mg/dL) 94.0 � 15.7 124.1 � 16.4**

*p < 0.05,**p < 0.01 vs. control by t-test.

ARTICLE IN PRESS

Endothelial function and arterial stiffness 3

+ MODEL

Endothelial function

Endothelial function was measured 1 h before the OGTTwas started. All studies were performed in the morning ina quiet room with a temperature of approximately 22 �C.

Brachial artery reactivity was measured by Dopplerultrasonography, using a 7.5 MHz linear array transducer ul-trasound system (ATL HDI 3000) on a 1e2 cm segment of theright artery located 2e4 cm above the antecubital crease.After the baseline image of brachial artery diameter hadbeen obtained, limb ischemia was produced by inflatinga standard sphygmomanometry cuff on the upper arm to200 mmHg for 5 min. After the cuff was deflated, brachialartery diameter was recorded at 60 s and 90 s during the hy-peremic response as an index of endothelium-dependentvasodilation. The ECG was monitored continuously through-out the study. After a 15-min recovery period, baseline re-cordings of brachial arterial diameter were repeated withsubsequent administration of 0.4 mg sublingual nitroglyc-erin (NTG). For the next 4 min, brachial arterial diameterwas continuously monitored to assess non-endothelium-dependent dilation. NTG serves as an exogenous source ofNO, bypassing the need for endogenous endothelial NO pro-duction. Brachial artery diameter at end diastole, definedby the R wave on the ECG, was subsequently analyzed off-line from the video recording by a single blinded observer,using electronic calipers. At each time point, the arterialdiameter was determined from the mean of three diametermeasurements for each of five cardiac cycles. Percentagechanges in vessel diameter in response to reactive hyper-emia and NTG were determined by dividing the differencebetween the maximal and the baseline diameters by thebaseline diameter. In our laboratory the coefficient of var-iation for repeated measurements of brachial artery diam-eter was 3.4%.

Assessment of pulse wave velocity (PWV)

Carotid-femoral PWV was determined by an automaticdevice (Complior�, Colson). The validation of this auto-matic method and its reproducibility have been previouslypublished [19]. Pulse transit time was determined as theaverage of 10 consecutive beats. The distance traveled bythe pulse wave was measured over the body surface asthe distance between the two recording sites (carotid andfemoral pulse). PWV was calculated as the ratio of distanceto transit time.

Repeated measurements were taken in 20 subjects (6males and 14 females). Correlation between repeatedmeasurements was 0.92. The mean difference betweenthe two measurements was 0.08 m/s (p Z 0.73). Coeffi-cient of variation was 1.4%.

Circulating inflammatory markers

At the beginning of each study day, fasting blood sampleswere taken. Multiple aliquots of serum were stored at�80 �C until analysis. Samples were frozen and thawed onlyonce to evaluate hsCRP, CD40L, TNF-a, IL-1b, VCAM-1, andICAM using commercially available kits. Serum hsCRP levelswere measured using a nephelometric assay (Dade-Behring,

Please cite this article in press as: Scuteri A et al., Endothelial functionrelatives of diabetic patients are independent of the metabolicj.numecd.2007.03.008

Liederbach, Germany). CD40L, IL-1b, VCAM-1, ICAM andTNF-a levels were evaluated by ELISA (Bender MedSystem)and assessed in duplicate.

Statistical analysis

All analyses were performed using the SPSS 8.0 software.Data are presented as mean � SD unless otherwise speci-fied. A normal distribution was observed for all the studiedparameters, except insulin and triglyceride levels.

Differences in mean values for each of the measuredvariables were compared by t-test for continuous variablesand by a chi-squared test for categorical variables. Compar-ison of different ‘‘insulin resistant’’ groups was made byANOVA, followed by Bonferroni’s test for multiple compar-isons. An ANCOVA analysis was used to compare groups af-ter adjustment for age, sex, or other covariates. Univariatelinear regression analysis was carried out by Pearson’s cor-relation or Spearman’s rank, as appropriate. Multiple linearregression analysis, with backward elimination, was used todetect statistically significant associations between FDRstatus and endothelial function or arterial stiffness, con-trolling for a number of independent variables: FDR status,age, female sex, current smoking, SBP, DBP, waist circum-ference, total cholesterol, HDL cholesterol, triglycerides,fasting glucose, HOMA-IR, ICAM, VCAM, sCD40L. A p valueof <0.05 was considered statistically significant.

Results

The clinical and vascular data of FDR and the control groupare shown in Table 1. There was no statistical difference inage, sex, and current smoking status between the twogroups. FDR presented higher BMI, waist circumference,fasting and 2 h post OGTT glucose and insulin levels, triglyc-erides and lower HDL cholesterol when compared to con-trol. Insulin sensitivity, evaluated by HOMA IR, was lower

and arterial stiffness in normotensive normoglycemic first-degreesyndrome, Nutr Metab Cardiovasc Dis (2007), doi:10.1016/

Table 2 Results in the study population by FDR status

Controls First-degree

n 16 29Fasting insulin 7.3 � 2.4 10.1 � 6.0**2 h OGTT insulin 34.6 � 17.4 79.4 � 65.6**HOMA IR 1.56 � 0.49 2.53 � 1.46**

BA diameter (mm) 3.3 � 0.5 3.6 � 0.8FMD (%) 16.2 � 7.6 9.8 � 5.2**NTGMD (%) 17.5 � 7.1 15.2 � 5.4PWV (m/s) 9.2 � 1.2 9.6 � 1.2

HsCRP (mg/L) 0.68 � 0.39 1.64 � 1.96*IL-1b (pg/mL) 0.62 � 0.77 1.16 � 1.30*TNF-a (pg/mL) 6.78 � 3.30 9.46 � 2.84*ICAM (ng/mL) 279 � 53 270 � 78VCAM (ng/mL) 932 � 423 1003 � 321sCD40L (ng/mL) 3.6 � 5.1 10.0 � 8.3*

*p < 0.05,**p < 0.01 vs. control by t-test.

ARTICLE IN PRESS

4 A. Scuteri et al.

+ MODEL

in FDR than in the control group. No significant differencesin any blood measures or in any blood pressure measure wasobserved between the two groups. After age and sexadjustment, differences in BMI, waist circumference, 2 hpost OGTT glucose and insulin levels remained significant.

FDR (Table 2) showed a significantly impaired endothe-lial function as indicated by a 33% lower flow-mediated di-lation than the control group (9.8 � 5.2% vs. 16.2 � 7.6%,

FDR-IRFDR-ISC

FMD

(%)

25

20

15

10

5

0

p < 0.01

FDR-IRFDR-ISC

sCD

40L

(ng/

ml)

30

25

20

15

10

5

0

p < 0.01

Figure 1 Effects of insulin resistance on endothelial function (Fligand). C, controls; FDR-IS, FDR insulin sensitive; FDR-IR, FDR insulin

Please cite this article in press as: Scuteri A et al., Endothelial functionrelatives of diabetic patients are independent of the metabolicj.numecd.2007.03.008

p < 0.01). PWV was 5% higher in FDR than in controls, butthis difference did not attain statistical significance. Circu-lating inflammatory markers, hsCRP, IL-1b, and TNF-a, weresignificantly higher in FDR than in controls. Serum CD40ligand was significantly higher in FDR than in controls(10.0 � 8.3 vs. 3.6 � 5.1 ng/mL, p < 0.01). After age andsex adjustment, differences in FMD, in TNF-a, and in serumCD40L remained significant.

Of note, when the upper quartile of HOMA IR was used todefine insulin-resistant subjects, no control and 11 (or 38%)of FDR showed insulin resistance. Insulin resistant subjectswere older and did not present any significant difference inblood pressure levels, whereas obesity, glucose and lipidlevels were higher than in insulin sensitive subjects, controlor FDR. As shown in Fig. 1, FMD was reduced in FDR, with orwithout insulin resistance. Conversely, arterial stiffnesswas significantly increased only in FDR with insulin resis-tance. Insulin resistance was also accompanied by proin-flammatory status, even in the absence of hyperglycemia.To further confirm the observed independency betweenFMD and insulin resistance, FMD was lower in FDR than incontrols even after adjustment for age, sex, and HOMA IR(9.7 � 1.1 vs. 14.8 � 1.5, p Z 0.011).

Univariate regression analysis revealed that FMD wasinversely correlated with FDR status, waist circumferenceand triglycerides, and positively correlated with female sexand HDL cholesterol; PWV was negatively correlated withfemale sex and positively correlated with mean bloodpressure levels, and insulin ‘‘milieu’’. Multiple regressionanalysis showed that FDR status was the strongest

FDR-IRFDR-ISC

PWV

(m/s

ec)

12,0

11,0

10,0

9,0

8,0

p < 0.05

FDR-IRFDR-ISC

HO

MA

IR

6

5

4

3

2

1

p < 0.001

MD), arterial stiffness (PWV), proinflammatory markers (CD40resistant. p value refers to ANOVA analysis for the three groups.

and arterial stiffness in normotensive normoglycemic first-degreesyndrome, Nutr Metab Cardiovasc Dis (2007), doi:10.1016/

Table 3 Determinants of FMD and PWV at multiple regres-sion analysis

Beta SE p

(a) FMDConstant 4.98FDR status �8.30 0.90 0.000Female sex 6.74 0.87 0.000SBP �0.27 0.08 0.008DBP 0.34 0.10 0.007Total cholesterol 0.049 0.012 0.003Current smoking �3.71 1.19 0.011ICAM 0.015 0.006 0.037

T(b) PWV(Constant) 16.554Fasting glucose 0.129 0.044 0.012sCD40L 0.112 0.043 0.023Total cholesterol 0.028 0.011 0.028

Model included FDR status, age, female sex, current smoking,SBP, DBP, waist circumference, total cholesterol, HDL choles-terol, triglycerides, fasting glucose, HOMA-IR, ICAM, VCAM,and sCD40L as independent variables.

Table 4 Comparison of subjects with no altered compo-nents of the metabolic syndrome according to FDR status(means � SE)

Controls First-degree

n 12 17Current smoker (%) 17 23Waist (cm) 80.3 � 8.5 82.3 � 8.3SBP (mmHg) 110 � 9 110 � 10DBP (mmHg) 74 � 8 71 � 8Total cholesterol (mg/dL) 174.3 � 23.3 179.6 � 32.0LDL cholesterol (mg/dL) 105.3 � 26.7 114.1 � 36.1HDL cholesterol (mg/dL) 71.1 � 12.9 64.0 � 14.8Triglycerides (mg/dL) 70.6 � 17.6 78.1 � 29.9Fasting glucose (mg/dL) 87.4 � 5.7 91.8 � 10.22 h OGTT glucose (mg/dL) 97.2 � 7.1 122.3 � 10.9*Fasting insulin 7.1 � 2.2 8.5 � 4.72 h OGTT insulin 38.2 � 9.0 59.6 � 7.4HOMA IR 1.62 � 0.28 1.94 � 0.23

FMD (%) 16.8 � 2.0 11.2 � 1.6*NTGMD (%) 18.2 � 2.0 17.2 � 0.7PWV (m/s) 8.8 � 0.3 9.6 � 0.3*

HsCRP (mg/L) 0.63 � 0.36 1.89 � 3.30IL-1b (pg/mL) 0.66 � 0.66 1.76 � 1.25*TNF-a (pg/mL) 6.69 � 3.15 7.59 � 4.70ICAM (ng/mL) 286 � 57 284 � 55VCAM (ng/mL) 1014 � 419 991 � 304sCD40L (ng/mL) 4.3 � 2.6 10.4 � 2.4*

*p < 0.05 vs. control.

ARTICLE IN PRESS

Endothelial function and arterial stiffness 5

+ MODEL

determinant of FMD, after controlling for FDR status, age,sex, current smoking, blood pressure, waist circumference,lipids, fasting glucose, HOMA-IR, ICAM, VCAM, and sCD40L(Table 3). In a similar multiple regression model, fastingglucose and sCD40 ligand persisted as significant indepen-dent determinants of PWV (Table 3).

Endothelial function and arterial stiffness: Roleof FDR status and of the metabolic syndrome

FDR subjects showed higher values of factors contributingto the metabolic syndrome (Table 1). Therefore, we triedto disentangle the role of FDR status per se from the roleof altered components of the metabolic syndrome. Indeed,the prevalence of altered components of the metabolicsyndrome and the number of subjects with one or more al-tered components of the metabolic syndrome differed be-tween FDR and control subjects. Specifically, 7 controlsubjects presented one altered component and nonemore than one altered component of the metabolic syn-drome; in the FDR group, 4 subjects presented one alteredcomponent, 4 subjects two altered components, and 4 sub-jects three or more altered components of the metabolicsyndrome. Given the small number of subjects with alteredcomponents of the metabolic syndrome in the controlgroup, we compared subjects with no altered componentsof the metabolic syndrome according to their FDR status.Table 4 shows that FDR subjects with no altered compo-nents of the metabolic syndrome presented a blunted endo-thelial function (lower FMD) and stiffer large arteries(higher PWV) than controls. Additionally, they presenteda proinflammatory status with higher levels of cytokines,specifically IL-1b and sCD40 ligand.

The observed differences in PWV remained significantwhen PWV was normalized for mean blood pressure andafter further adjustment for age and sex.

Please cite this article in press as: Scuteri A et al., Endothelial functionrelatives of diabetic patients are independent of the metabolicj.numecd.2007.03.008

Discussion

The present study showed that: (a) normotensive normo-glycemic first-degree relatives of type 2 diabetic subjectspresent early endothelial dysfunction, even in the absenceof insulin resistance; (b) large arteries stiffening (increasedPWV) is paralleled by decreased insulin sensitivity beforehyperglycemia or blood pressure elevation become evident;(c) alteration of endothelial function and arterial stiffeningobserved in normotensive normoglycemic first-degree rel-atives of type 2 diabetic subjects is independent of themetabolic syndrome.

Previous studies investigated endothelial function inFDR, but no study investigated arterial stiffness in FDR.Ballesthofer et al. showed no significant difference in FMDbetween FDR and controls; these differences becamesignificant only in subjects with insulin resistance [20].Our findings confirm these results, although our control sub-jects presented a higher FMD as compared to that describedby Ballesthofer [20]. We cannot rule out that such differ-ences may be, at least partly, attributable to differenttechniques used to evaluate endothelial function. In fact,Ballesthofer used lower cuff occlusion, whereas we usedupper cuff occlusion. However, in the currently publishedguidelines both techniques are considered to adequatelyreflect brachial artery endothelial function [10]. An alter-native possible explanation of different findings betweenthe two studies is the very high prevalence of current

and arterial stiffness in normotensive normoglycemic first-degreesyndrome, Nutr Metab Cardiovasc Dis (2007), doi:10.1016/

ARTICLE IN PRESS

6 A. Scuteri et al.

+ MODEL

smoking in the Ballesthofer study, given that smoking isknown to damage endothelial function [21,22]. Our findingsare in agreement with Caballero et al. [23] who showed de-creased FMD (and decreased microvascular reactivity) inFDR as compared to controls. It should be acknowledgedthat their population was on average 20 years older thanours. Additionally, in contrast to our study, no significantdifference in the levels of the metabolic syndrome compo-nents was observed between controls and FDR [23].

Although our study population consisted of normotensivenormoglycemic first-degree relatives of diabetic subjects,they presented metabolic alterations and increased prev-alence of altered components of the metabolic syndrome.Both first-degree status and the presence of the metabolicsyndrome [24] are associated with increased risk of devel-oping diabetes mellitus. The metabolic syndrome hasbeen associated with alteration in vascular structure andfunction, namely thickening and stiffening of large arteries[15,25e29]. The metabolic syndrome and its componentshave also been associated with endothelial dysfunction[30,31]. Our findings suggest that the endothelial dysfunc-tion observed in normotensive normoglycemic first-degreerelatives of diabetic subjects is detectable even in the ab-sence of any altered components of the metabolic syn-drome. Pulse wave velocity is also mildly increased innormotensive normoglycemic first-degree relatives of dia-betic subjects with no altered components of the metabolicsyndrome, but arterial stiffening is more evident in thepresence of insulin-resistance.

The cross-sectional design of the present study does notallow inference on causality. However, we can speculatethat normotensive normoglycemic first-degree relatives oftype 2 diabetic patients present a blunted endothelialfunction in relation to a primary inherited genetic suscep-tibility yet to be identified, or in relation to the metabolicalteration detectable even in normotensive and normogly-cemic subjects. Certainly, the blunted endothelial functionobserved in FDR subjects may also be related to elevatedglucose levels, given that hyperglycemia has been shown todecrease NO bioavailability through activation of PKC viadiacylglycerol, increased hexosamine pathway flux, in-creased advanced glycation end product formation, andincreased polyol pathway flux [32e34]. Indeed, our FDRsubjects had still normal but already higher glycemia ascompared to controls. Proinflammatory status (increasedcirculating levels of hsCRP, IL-1b, TNF-a, and CD40 ligand)further deteriorate ability of endothelium to restore itsfunction [35]. Both endothelial dysfunction and the proin-flammatory status progressively impair insulin sensitivity,before any alteration in fasting plasma glucose can be de-tecteddas suggested by a study where inhibition of nitricoxide by L-NNMA not only impaired endothelial function,but also reduced insulin sensitivity by impairing skeletalmuscle glucose uptake [36]. The cross-talk between proin-flammatory status, endothelial dysfunction, and insulinsensitivity [13] further deteriorate large artery stiffnessand will maintain a vicious circle responsible for the evolu-tion of the metabolic syndrome, incident diabetes mellitusand cardiovascular disease, frequent complications in thenatural history of type 2 diabetes. Microcirculation mayalso represent the first vascular area to suffer frompremature endothelial cell damage/senescence [37].

Please cite this article in press as: Scuteri A et al., Endothelial functionrelatives of diabetic patients are independent of the metabolicj.numecd.2007.03.008

Microcirculation damage may progressively lead to in-creased large artery stiffness [38,39], possibly throughproximalization of pulse wave reflection site [40], prior toblood pressure increase. Indeed, the association betweenmicrovascular narrowing and large artery stiffness hasbeen described in healthy middle-aged men and women[41].

To what extent deterioration in insulin sensitivity con-tributes to stiffen arteries (increased PWV) in FDR in-dependently of endothelial dysfunction is difficult todisentangle. Indeed, we observed that FDR with no alteredcomponents of the metabolic syndrome presented endo-thelial dysfunction in stiffer large arteries. Consistentlywith this observation, recent studies have demonstratedthat endothelium is also involved in the regulation of PWV[42e44]. Kinlay et al. showed that inhibition of NO releaseby L-NNMA increased PWV, whereas nitroglycerin adminis-tration decreased PWV in young healthy men [42]. Simi-larly, in sheep, Wilkinson et al. demonstrated that basalNO production influenced large artery stiffness [43]. Stew-art et al. [44] showed that inhibition of basal NO synthesisby L-NMMA increased carotid-femoral PWV, but suggestedthat the change in mean blood pressure accounted for thechanges in PWV, given that the increase in carotid-femoralPWV was similar to that produced by equipressor doses ofnorepinephrine and dobutamine. On the other hand, thepresent study showed that increased PWV was accompaniedby insulin resistance in FDR, even after adjustment for sex.The Hoorn Study showed that arterial stiffness increasedbefore the onset of type 2 diabetes mellitus, even insubjects with impaired glucose metabolism, and it wasnot entirely explained by indices of hyperglycemia orhyperinsulinemia [45]. In contrast with our study, however,increased stiffness was accompanied by higher blood pres-sure levels [45]. Wright reported that in diabetic patients,PWV increased with abnormal glucose tolerance and dura-tion of diabetes [46]. We have previously reported thatthe metabolic syndrome was associated with increasedarterial stiffness in both men and women across differentage groups and was itself an independent predictor of stiff-ness [15].

In conclusion, normoglycemic first-degree relatives ofdiabetic subjects have blunted endothelial function andincreased stiffness of the large arteries. These alterationsare already present at a very young age, before anyalteration in glycemic control or blood pressure valuescan be detected. Such alterations are independent of thepresence of the metabolic syndrome and its alteredcomponents, although deterioration of insulin sensitivitycontributed to accelerate stiffening of large arteries.

References

[1] Kannel WB, McGee DL. Diabetes and cardiovascular disease.The Framingham study. JAMA 1979;241:2035e8.

[2] Pyorala K, Laakso M, Uusitupa M. Diabetes and athero-sclerosis: an epidemiologic view. Diabetes Metab Rev 1987;3:463e524.

[3] Kawamori R, Yamasaki Y, Matsushima H, Nishizawa H, Nao K,Hougaku H, et al. Prevalence of carotid atherosclerosis indiabetic patients. Ultrasound high-resolution B-mode imagingon carotid arteries. Diabetes Care 1992;15:1290e4.

and arterial stiffness in normotensive normoglycemic first-degreesyndrome, Nutr Metab Cardiovasc Dis (2007), doi:10.1016/

ARTICLE IN PRESS

Endothelial function and arterial stiffness 7

+ MODEL

[4] Uusitupa M, Niskanen L, Siitonen O, Pyorala K. Hyperinsuline-mia and hypertension in patients with newly diagnosed non-insulin-dependent diabetes. Diabetes Metab Rev 1987;13(3 Pt 2):369e74.

[5] Luscher TF. Mack Forster Award Lecture. Review. The endo-thelium as a target and mediator of cardiovascular disease.Eur J Clin Invest 1993;23:670e85.

[6] Suwaidi JA, Hamasaki S, Higano ST, Nishimura RA,Holmes Jr DR, Lerman A. Long-term follow-up of patientswith mild coronary artery disease and endothelial dysfunc-tion. Circulation 2000;101:948e54.

[7] Boutouyrie P, Tropeano AI, Asmar R, Gautier I, Benetos A,Lacolley P, et al. Aortic stiffness is an independent predictorof primary coronary events in hypertensive patients: a longitu-dinal study. Hypertension 2002;39:10e5.

[8] Meaume S, Benetos A, Henry OF, Rudnichi A, Safar ME. Aorticpulse wave velocity predicts cardiovascular mortality in sub-jects >70 years of age. Arterioscler Thromb Vasc Biol 2001;21:2046e50.

[9] Najjar SS, Scuteri A, Lakatta EG. Arterial aging: is it animmutable cardiovascular risk factor? Hypertension 2005;46:454e62.

[10] Corretti MC, Anderson TJ, Benjamin EJ, Celermajer D,Charbonneau F, Creager MA, et al. International Brachial Ar-tery Reactivity Task Force. Guidelines for the ultrasoundassessment of endothelial-dependent flow-mediated vasodila-tion of the brachial artery: a report of the International Bra-chial Artery Reactivity Task Force. J Am Coll Cardiol 2002;39:257e65.

[11] Behrendt D, Ganz P. Endothelial function. From vascular biol-ogy to clinical applications. Am J Cardiol 2002;90:40Le8.

[12] Ross R. The pathogenesis of atherosclerosis: a perspective forthe 1990s. Nature 1993;362:801e9.

[13] Hsueh WA, Lyon CJ, Quinones MJ. Insulin resistance and theendothelium. Am J Med 2004;117:109e17.

[14] Scuteri A, Najjar SS, Muller DC, Andres R, Hougaku H,Metter EJ, et al. Metabolic syndrome amplifies the age-associ-ated increases in vascular thickness and stiffness. J Am CollCardiol 2004;43:1388e95.

[15] Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA,Izzo Jr JL, et al. Joint National Committee on Prevention,Detection, Evaluation, and Treatment of High Blood Pressure.National Heart, Lung and Blood Institute; National High BloodPressure Education Program Coordinating Committee. Seventhreport of the Joint National Committee on Prevention, Detec-tion, Evaluation, and Treatment of High Blood Pressure.Hypertension 2003;42:1206e52.

[16] American Diabetes Association. Clinical practice recommen-dations 1997. Diabetes Care 1997;20(Suppl. 1):S1e70.

[17] Bonora E, Targher G, Alberiche M, Bonadonna RC, Saggiani F,Zenere MB, et al. Homeostasis model assessment closely mir-rors the glucose clamp technique in the assessment of insulinsensitivity: studies in subjects with various degrees of glucosetolerance and insulin sensitivity. Diabetes Care 2000;23:57e63.

[18] Executive Summary of The Third Report of The National Cho-lesterol Education Program (NCEP) Expert Panel on Detection,Evaluation And Treatment of High Blood Cholesterol In Adults(Adult Treatment Panel III). Expert Panel on Detection, Evalu-ation, and Treatment of High Blood Cholesterol in Adults.JAMA 2001;285:2486e97.

[19] Asmar R, Benetos A, Topouchian J, Laurent P, Pannier B,Brisac AM, et al. Assessment of arterial distensibility by auto-matic pulse wave velocity measurement: validation and clini-cal application studies. Hypertension 1995;26:485e90.

[20] Balletshofer BM, Rittig K, Enderle MD, Volk A, Maerker E,Jacob S, et al. Endothelial dysfunction is detectable in youngnormotensive first-degree relatives of subjects with type 2

Please cite this article in press as: Scuteri A et al., Endothelial functionrelatives of diabetic patients are independent of the metabolicj.numecd.2007.03.008

diabetes in association with insulin resistance. Circulation2000;101:1780e4.

[21] Celermajer DS, Sorensen KE, Bull C, Robinson J, Deanfield JE.Endothelium dependent dilation in the systemic arteries ofasymptomatic subjects relates to coronary risk factors andtheir interaction. J Am Coll Cardiol 1994;24:1468e74.

[22] Celermajer DS, Adams MR, Clarkson P, Robinson J, McCredie R,Donald A, et al. Passive smoking and impaired endotheliumdependent arterial dilation in healthy young adults. N Engl JMed 1996;334:150e4.

[23] Caballero AE, Arora S, Saouaf R, Lim SC, Smakowski P,Park JY, et al. Microvascular and macrovascular reactivity isreduced in subjects at risk for type 2 diabetes. Diabetes1999;48:1856e62.

[24] Laaksonen D, Lakka H-M, Niskanen L, Kaplan G, Salonen J,Lakka T. Metabolic syndrome and development of diabetesmellitus: application and validation of recently suggested def-initions of the metabolic syndrome in a prospective cohortstudy. Am J Epidemiol 2002;156:1070e7.

[25] Czernichow S, Bertrais S, Blacher J, Oppert JM, Galan P,Ducimetiere P, et al. Metabolic syndrome in relation to struc-ture and function of large arteries: a predominant effect ofblood pressure. A report from the SU.VI.MAX. Vascular Study.Am J Hypertens 2005;18(9 Pt 1):1154e60.

[26] Anand SS, Yi Q, Gerstein H, Lonn E, Jacobs R, Vuksan V, et al.Relationship of metabolic syndrome and fibrinolytic dysfunc-tion to cardiovascular disease. Circulation 2003;108:420e5.

[27] Hedblad B, Nilsson P, Janzon L, Berglund G. Relation betweeninsulin resistance and carotid intima-media thickness and ste-nosis in non-diabetic subjects. Results from a cross-sectionalstudy in Malmo, Sweden. Diabet Med 2000;17:299e307.

[28] Hulthe J, Bokemark L, Wikstrand J, Fagerberg B. The meta-bolic syndrome, LDL particle size, and atherosclerosis: theAtherosclerosis and Insulin Resistance (AIR) study. ArteriosclerThromb Vasc Biol 2000;20:2140e7.

[29] van Popele NM, Westendorp IC, Bots ML, Reneman RS,Hoeks AP, Hofman A, et al. Variables of the insulin resistancesyndrome are associated with reduced arterial distensibility inhealthy non-diabetic middle-aged women. Diabetologia 2000;43:665e72.

[30] Toikka JO, Ahotupa M, Viikari JSA, Niinikoski H, Taskinen M,Iriala K, et al. Constantly low HDL cholesterol concentrationrelates to endothelial dysfunction and increased in vivo LDL-oxidation in healthy young men. Atherosclerosis 1999;147:133e8.

[31] Matsuoka H. Endothelial dysfunction associated with oxidativestress in humans. Diabetes Res Clin Pract 2001;54(suppl.):S65e72.

[32] Stepp DW, Ou J, Ackerman AW, Welak S, Klick D,Pritchard Jr KA. Native LDL and minimally oxidized LDL differ-entially regulate superoxide anion in vascular endothelium insitu. Am J Physiol Heart Circ Physiol 2002;283:H750e9.

[33] Zou MH, Shi C, Cohen RA. Oxidation of the zinc-thiolate com-plex and uncoupling of endothelial nitric oxide synthase byperoxynitrite. J Clin Invest 2002;109:817e26.

[34] Guzik TJ, Mussa S, Gastaldi D, Sadowski J, Ratnatunga C,Pillai R, et al. Mechanisms of increased vascular superoxideproduction in human diabetes mellitus: role of NAD(P)H oxi-dase and endothelial nitric oxide synthase. Circulation 2002;105:1656e62.

[35] Urbich C, Dernbach E, Aicher A, Zeiher AM, Dimmeler S. CD40ligand inhibits endothelial cell migration by increasing produc-tion of endothelial reactive oxygen species. Circulation 2002;106:981e6.

[36] Baron AD, Steinberg HO, Chaker H, Leaming R, Johnson A,Brechtel G. Insulin-mediated skeletal muscle vasodilationcontributes to both insulin sensitivity and responsiveness inlean humans. J Clin Invest 1995;96:786e92.

and arterial stiffness in normotensive normoglycemic first-degreesyndrome, Nutr Metab Cardiovasc Dis (2007), doi:10.1016/

ARTICLE IN PRESS

8 A. Scuteri et al.

+ MODEL

[37] Brodsky SV, Gealekman O, Chen J, Zhang F, Togashi N,Crabtree M, et al. Prevention and reversal of premature endo-thelial cell senescence and vasculopathy in obesity-induceddiabetes by ebselen. Circ Res 2004;94:377e84.

[38] Sullivan JM, Prewitt RL, Josephs JA. Attenuation of the micro-circulation in young patients with high-output borderline hy-pertension. Hypertension 1983;5:844e51.

[39] Isaksson H, Cederholm T, Jansson E, Nygren A, Ostergren J.Therapy-resistant hypertension associated with central obe-sity, insulin resistance, and large muscle fibre area. BloodPress 1993;2:46e52.

[40] O’Rourke MF, Kelly RP. Wave reflection in the systemic circu-lation and its implications in ventricular function. J Hypertens1993;11:327e37.

[41] Liao D, Wong TY, Klein R, Jones D, Hubbard L, Sharrett AR. Re-lationship between carotid artery stiffness and retinal arteri-olar narrowing in healthy middle-aged persons. Stroke 2004;35:837e42.

Please cite this article in press as: Scuteri A et al., Endothelial functionrelatives of diabetic patients are independent of the metabolicj.numecd.2007.03.008

[42] Kinlay S, Creager MA, Fukumoto M, Hikita H, Fang JC,Selwyn AP, et al. Endothelium-derived nitric oxide regulatesarterial elasticity in human arteries in vivo. Hypertension2001;38:1049e53.

[43] Wilkinson IB, Qasem A, McEniery CM, Webb DJ, Avolio AP,Cockcroft JR. Nitric oxide regulates local arterial distensibilityin vivo. Circulation 2002;105:213e7.

[44] Stewart AD, Millasseau SC, Kearney MT, Ritter JM,Chowienczyk PJ. Effects of inhibition of basal nitric oxide syn-thesis on carotid-femoral pulse wave velocity andaugmentation index in humans. Hypertension 2003;42:915e8.

[45] Henry RM, Kostense PJ, Spijkerman AM, Dekker JM, Nijpels G,Heine RJ, et al. Hoorn Study. Arterial stiffness increases withdeteriorating glucose tolerance status: the Hoorn Study.Circulation 2003;107:2089e95.

[46] Wright JS. Doppler ultrasound measurement of aortic compli-ance in diabetic patients and community control. PhD thesis.London: University of London; 1991.

and arterial stiffness in normotensive normoglycemic first-degreesyndrome, Nutr Metab Cardiovasc Dis (2007), doi:10.1016/