at SciVerse ScienceDirect

Atherosclerosis 229 (2013) 453e458

Contents lists available

Atherosclerosis

journal homepage: www.elsevier .com/locate/atherosclerosis

Insulin clearance is associated with carotid artery intimaemediathickness

Maria A. Marini a, Simona Frontoni a, Elena Succurro b, Franco Arturi b,Teresa V. Fiorentino b, Angela Sciacqua b, Marta L. Hribal b, Francesco Perticone b,Giorgio Sesti b,*aDepartment of Systems Medicine, University of Rome-Tor Vergata, Rome, ItalybDepartment of Medical and Surgical Sciences, University “Magna Graecia” of Catanzaro, Viale Europa, 88100 Catanzaro, Italy

a r t i c l e i n f o

Article history:Received 25 April 2013Received in revised form13 June 2013Accepted 13 June 2013Available online 22 June 2013

Keywords:Insulin clearanceCarotid artery intimaemedia thicknessVascular damageNon-diabetic subjects

* Corresponding author. Tel.: þ39 0961 3647204; faE-mail address: sesti@unicz.it (G. Sesti).

0021-9150/$ e see front matter � 2013 Elsevier Irelahttp://dx.doi.org/10.1016/j.atherosclerosis.2013.06.011

a b s t r a c t

Objective: The aim of this study was to investigate whether insulin clearance is independently associatedwith carotid artery intimaemedia thickness (IMT), a well-recognized index of vascular damage.Methods: 361 Non-diabetic Caucasian subjects were subjected to euglycemic hyperinsulinemic clamp toassess insulin sensitivity, and insulin clearance. IMT of the common carotid was measured byultrasonography.Results: Among the study group, 270 subjects had normal glucose tolerance, 33 had impaired fastingglucose, and 58 had impaired glucose tolerance. Univariate correlations showed that age, BMI, waist, bloodpressure, triglycerides, fasting and 2-h post-load glucose and insulin levels were positively correlated withcarotid IMT whereas HDL, insulin clearance, and insulin-stimulated glucose disposal were negativelycorrelated with IMT. A multivariate regression analysis in a model including, in addition to insulin clear-ance, age, gender, BMI, waist, blood pressure, triglycerides, HDL, fasting and 2-h post-load glucose, insulin-stimulated glucose disposal, fasting and 2-h post-load insulin showed that the traits independentlyassociated with carotid IMT were BMI (b ¼ 0.42, P < 0.0001), insulin clearance (b ¼ �0.29, P < 0.0001), age(b ¼ 0.19, P < 0.0001), waist (b ¼ 0.18, P ¼ 0.01), diastolic blood pressure (b ¼ 0.17, P ¼ 0.01), and 2-h post-load glucose (b ¼ 0.12, P ¼ 0.03). These factors explained 26% of the variance in carotid IMT. Subjects in thelowest tertile of insulin clearance had a 4.06-fold higher odds of having vascular damage (IMT> 0.9 mm) ascompared with those in the highest tertile (OR 4.06, 95%CI 1.15e13.24).Conclusions: Insulin clearance is independently associated with carotid IMT in adult non-diabetic sub-jects. Individuals with lower levels of insulin clearance have a higher odds of vascular damage.

� 2013 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

Insulin resistance is associated to several conditions such as type2 diabetes mellitus [1,2], and cardiovascular disease [3,4], and maybe part of the causal pathway that links these conditions to car-diovascular risk factors, including visceral obesity, atherogenicdyslipidemia, and hypertension, clustering within the metabolicsyndrome [5,6]. Hyperinsulinemia, which is a surrogate measure ofinsulin resistance, may also affect the development of vascularatherosclerosis, indeed an association between elevated fastinginsulin levels and the risk of atherosclerotic cardiovascular diseasehas been reported by many [7e10] but not all [11e13] prospective

x: þ39 0961 3647192.

nd Ltd. All rights reserved.

studies. Plasma insulin concentrations are influenced by distinctphysiological processes, including insulin sensitivity, insulinsecretion, and insulin clearance. It has been demonstrated thatinsulin clearance is the strongest determinant of the variability infasting insulin levels independently of insulin resistance, insulinsecretion, adiposity, and fasting plasma glucose [14]. On the otherhand, decreased insulin clearance has been reported in conditionsof insulin resistance [15e19], and predicts the development of type2 diabetes independently of confounding factors [20]. To the best ofour knowledge, no studies have investigated the association ofsubclinical atherosclerosis to insulin clearance. Particularly, theextent to which subclinical atherosclerosis is due to insulin resis-tance, per se, versus insulin clearance is still undefined. Exploringthis relationship is important in order to gain insight into potentialpathophysiological mechanisms underlying the development ofcardiovascular disease. The study of the relative role of insulin

Table 1Clinical characteristics of the study cohort.

Overall cohort Men Women P value

Number 361 173 188 e

Age (yrs) 39 � 10 41 � 10 37 � 10 <0.0001BMI (kg/m2) 29.7 � 7.5 29.4 � 5.6 29.9 � 8.9 0.52Waist circumference

(cm)95 � 15 100 � 12 91 � 17 <0.0001

SBP (mmHg) 126 � 17 131 � 14 120 � 17 <0.0001DBP (mmHg) 81 � 11 85 � 10 77 � 10 <0.0001Total cholesterol (mg/dl) 195 � 37 198 � 37 193 � 37 0.22HDL cholesterol (mg/dl) 52 � 14 45 � 10 59 � 14 <0.0001Triglyceride (mg/dl) 120 � 73 144 � 82 98 � 55 <0.0001Fasting glucose (mg/dl) 90 � 16 94 � 18 86 � 12 <0.00012-h Glucose (mg/dl) 114 � 33 118 � 34 111 � 32 0.057Fasting insulin (mU/ml) 12 � 8 11 � 7 13 � 9 0.152-h Insulin (mU/ml) 72 � 75 73 � 65 70 � 68 0.71NGT/IGT/IFG (n) 270/33/58 112/27/34 158/6/24 <0.0001Insulin-stimulated

glucose disposal(mg � kg�1 � min�1)

6.3 � 3.3 5.6 � 2.9 6.9 � 3.5 <0.0001

Insulin clearance(ml/min � m2)

530 � 235 523 � 180 537 � 275 0.57

Intimaemediathickness (mm)

0.74 � 0.12 0.77 � 0.10 0.73 � 0.13 0.01

Data are means � SD. M ¼ male; F ¼ female; SBP ¼ systolic blood pressure;DBP ¼ diastolic blood pressure; NGT ¼ normal glucose tolerance; IGT ¼ impairedglucose tolerance; IFG ¼ impaired fasting glucose.

M.A. Marini et al. / Atherosclerosis 229 (2013) 453e458454

clearance versus insulin resistance requires an independentassessment of the two phenomena and the use of multivariatestatistical models including confounding cardiovascular risk fac-tors. Intimaemedia thickness (IMT) of common carotid artery is awell-recognized index of vascular damage, and its increase predictsfuture development of myocardial infarction and stroke [21]. In thepresent study, we hypothesized that insulin clearance is indepen-dently associated with carotid IMT. To examine this, we havemeasured carotid artery IMT by ultrasonography in a cohort of non-diabetic Caucasian individuals in whom insulin sensitivity and in-sulin clearance were assessed by the euglycemic hyperinsulinemicclamp technique.

2. Methods

The study cohort comprised 361 non-diabetic offspringparticipating to the EUGENE2 project [22,23] who had only one ofthe parents with type 2 diabetes. All subjects were Caucasian andwere consecutively recruited at the Department of Systems Medi-cine of the University of Rome-Tor Vergata and at the Departmentof Medical and Surgical Sciences of the University “Magna Graecia”of Catanzaro. Subjects were excluded if they had history of car-diovascular disease including myocardial infarction, stoke, periph-eral atherosclerosis, chronic gastrointestinal diseases associatedwith malabsorption, chronic pancreatitis, history of any malignantdisease, history of alcohol or drug abuse, positivity for antibodies tohepatitis C virus (HCV) or hepatitis B surface antigen (HBsAg), andliver or kidney failure. All individuals were studied according to apreviously described protocol [22,24]. All subjects underwentanthropometrical evaluation including measurements of bodymass index (BMI), andwaist circumference. Readings of clinic bloodpressure (BP) were obtained in the left arm of the supine patients,after 5 min of quiet rest, with a sphygmomanometer. Values werecalculated as the average of the last two of three consecutivemeasurements obtained at 3-min intervals. After 12-h fasting, a75 g OGTT was performed with 0, 30, 60, 90 and 120 min samplingfor plasma glucose and insulin assay. Insulin sensitivity wasassessed by euglycemic hyperinsulinemic clamp study, as previ-ously described [25]. Briefly, a continuous insulin infusion wasinitiated at the rate of 40 mU/m2 of body surface area per min, aftera priming dose, in order to reach and maintain steady state plasmainsulin of 87 � 10 mU/ml. Plasma glucose was assessed at 5 minintervals during the 2-h clamp study by a glucose analyzer. In thestudy subjects, mean plasma glucose concentration during the lasthour of the clamp was 93 � 5 mg/dl. Plasma insulin concentrationswere measured every 20 min during the insulin infusion. The studywas approved by Institutional Ethics Committees and informedconsent was obtained from each subject in accordance with prin-ciples of the Declaration of Helsinki.

2.1. Ultrasound measurement of IMT of the common carotid artery

High resolution B-mode ultrasound was used to measure IMT ofthe common carotid artery by using an ATL HDI 3000 ultrasoundsystem (Advanced Technology Laboratories, Bothell, WA) equippedwith a 5 MHz linear array transducer. The thickness of the intimaemedia complex was assessed as previously described [24]. Manualmeasurements were conducted in plaque-free portions of the 10-mm linear segment proximal to the carotid bulb. For each subjecttwomeasurements were performed bilaterally and the values wereaveraged. Ultrasound measurement of IMT was performed by atrained examiner who was unaware of the subjects’ clinical andlaboratory findings. A value of IMT>0.9mmwas used as an index ofvascular damage according to the 2007 Guidelines for the man-agement of arterial hypertension released by the Task Force for the

Management of Arterial Hypertension of the European Society ofHypertension (ESH) and of the European Society of Cardiology(ESC) [26].

2.2. Calculation

Glucose disposal (M) was calculated as the mean rate of glucoseinfusion measured during the last 60 min of the clamp examination(steady-state) and is expressed as milligrams per minute per kilo-gram. Insulin clearance was calculated by dividing the rate of insulininfusion by the mean steady-state plasma insulin concentrationduring the euglycemic clamp, as previously described [25].

2.3. Statistical analyses

Variables with skewed distribution including triglycerides,fasting and 2-h post-load insulin were natural log transformed forstatistical analyses. Continuous variables are expressed asmeans � SD. Categorical variables were compared by c2 test. Re-lationships between variables were determined by Pearson’s cor-relation coefficient (r). The relationship between gender and IMTwas determined by pointebiserial correlation coefficient. Partialcorrelation coefficients adjusted for age and genderwere computedbetween variables. Relationships between variables were sought bymultivariate linear regression analysis in order to assess themagnitude of their individual effect on carotid IMT. A multivariablelogistic regression analysis was used to determine the associationbetween vascular damage and the tertiles of insulin clearance. Forall analyses a P value � 0.05 was considered to be statisticallysignificant. All analyses were performed using SPSS softwareVersion 16.0 for Windows.

3. Results

The clinical and metabolic characteristics of the study cohort(173 men, 188 women) are shown in Table 1. Among the studysubjects, 270 subjects (74.8%) had normal glucose tolerance (NGT),33 subjects (9.1%) had impaired fasting glucose (IFG), and 58

M.A. Marini et al. / Atherosclerosis 229 (2013) 453e458 455

subjects (16.1%) had impaired glucose tolerance (IGT). Men wereolder than women, and exhibited higher values of waist circum-ference, blood pressure, triglyceride, fasting glucose, IMT, and lowerHDL cholesterol, and insulin-stimulated glucose disposal.

Univariate correlations between carotid IMT and anthropo-metric and metabolic risk factors of atherosclerosis (Table 2)showed that gender, age, BMI, waist circumference, systolic bloodpressure (SBP), diastolic blood pressure (DBP), fasting and 2-h post-load glucose, fasting plasma and 2-h post-load insulin, and tri-glycerides levels were all positively correlated with carotid IMTwhereas HDL cholesterol, insulin clearance, and insulin-stimulatedglucose disposal, as a measure of insulin sensitivity, were nega-tively correlated with IMT. These correlations remained significantafter adjusting for gender and age with the exception of fastingplasma glucose (Table 2).

Univariate correlations between carotid IMT and anthropo-metric and metabolic risk factors of atherosclerosis stratified bygender are shown in Table 3. In the men sub-group, age, BMI, SBP,DBP, fasting and 2-h post-load glucose were positively correlatedwith carotid IMT whereas insulin clearance, and insulin-stimulatedglucose disposal were negatively correlated with IMT. BMI, DBP,insulin clearance, and insulin-stimulated glucose disposal were theonly variables that remained significantly associated with carotidIMTafter adjusting for age (Table 3). Inwomen sub-group, age, BMI,waist circumference, SBP, DBP, 2-h post-load glucose, fasting and 2-h post-load plasma insulin, and triglycerides levels were positivelycorrelated with carotid IMT whereas HDL cholesterol, insulinclearance, and insulin-stimulated glucose disposal were negativelycorrelated with IMT (Table 3). The correlation between insulinclearance and carotid IMT was stronger in women than in men asdetermined by testing the difference between the two correlationcoefficients (P ¼ 0.012).

After adjusting for age, the variables that remained significantlyassociated with carotid IMT were BMI, waist circumference, SBP,DBP, triglyceride, HDL cholesterol, 2-h post-load glucose, 2-h post-load plasma insulin, insulin clearance, and insulin-stimulatedglucose disposal.

To estimate the independent association of insulin clearance tocarotid IMT, we carried out a multivariate regression analysis in amodel including, in addition to insulin clearance, the metabolic

Table 2Univariate correlations between IMT and anthropometric and biochemical variables.

Unadjusted correlations betweenIMT and metabolic variables

Pearson’s correlationcoefficient (r)

Gender (male/female) 0.11Age (yrs) 0.26BMI (kg/m2) 0.30Waist circumference (cm) 0.26SBP (mmHg) 0.28DBP (mmHg) 0.28Total cholesterol (mg/dl) 0.09HDL cholesterol (mg/dl) �0.18Triglycerides (mg/dl) 0.26Fasting glucose (mg/dl) 0.142-h Glucose (mg/dl) 0.23Fasting insulin (mU/ml) 0.172-h Insulin (mU/ml) 0.14Insulin-stimulated glucose disposal

(mg � kg�1 � min�1)�0.17

Insulin clearance (ml/min � m2) �0.23

BMI ¼ body mass index; SBP ¼ systolic blood pressure; DBP ¼ diastolic blood pressure.results after analyses with adjustment for gender.

variables which were associated with carotid IMT in univariateanalysis, i.e. age, BMI, waist circumference, SBP, DBP, triglyceride,HDL cholesterol, fasting and 2-h post-load glucose, fasting and 2-hpost-load plasma insulin, insulin-stimulated glucose disposal andgender. Comparison of standardized coefficients allowed thedetermination of the relative strength of each trait’s associationwith carotid IMT (listed from strongest to weakest): BMI (b ¼ 0.42,P< 0.0001), insulin clearance (b¼�0.29, P< 0.0001), age (b¼ 0.19,P < 0.0001), waist circumference (b ¼ 0.18, P ¼ 0.01), DBP (b ¼ 0.17,P ¼ 0.01), and 2-h post-load glucose (b ¼ 0.12, P ¼ 0.03). Thesefactors explained 26% of the variance in carotid IMT.

Similar results were obtained when multivariate regressionanalysis was performed in the NGT group. Thus, the relativestrength of each trait’s association with carotid IMT was as follows:BMI (b ¼ 0.47, P < 0.0001), waist circumference (b ¼ 0.36,P< 0.0001), insulin clearance (b¼�0.31, P< 0.0001), age (b¼ 0.25,P ¼ 0.001), DBP (b ¼ 0.23, P ¼ 0.01), and fasting plasma glucose(b ¼ 0.15, P ¼ 0.03), These factors explained 26% of the variance incarotid IMT.

Next, the multivariate regression analyses were run separatelyin men and women in a model including the variables which wereassociated with carotid IMT in univariate analysis. In the men sub-group, the relative strength of each trait’s association with carotidIMT was as follows: BMI (b ¼ 0.37, P < 0.0001), DBP (b ¼ 0.33,P < 0.0001), age (b ¼ 0.24, P ¼ 0.005), and insulin clearance(b¼�0.15, P¼ 0.04). These factors explained 20% of the variance incarotid IMT. In the women sub-group, the relative strength of eachtrait’s association with carotid IMT was as follows: BMI (b ¼ 0.51,P < 0.0001), insulin clearance (b ¼ �0.38, P < 0.0001), waistcircumference (b ¼ 0.29, P ¼ 0.009), and triglycerides (b ¼ 0.13,P ¼ 0.05). These factors explained 33% of the variance in carotidIMT.

Then, thewhole study group subjects were stratified into tertilesaccording to the values of insulin clearance. In a logistic regressionmodel adjusted for age, gender, BMI, waist circumference, SBP, DBP,fasting and 2-h post-load glucose, fasting insulin, triglycerides, totalcholesterol, HDL cholesterol, and insulin-stimulated glucosedisposal, subjects in the lowest tertile of insulin clearance had a4.06-fold higher odds of having vascular damage as compared withthose in the highest tertile (OR 4.06, 95%CI 1.15e13.24).

P Age- and gender-adjustedcorrelations between IMTand metabolic variables

P

Pearson’s correlationcoefficient (r)

0.03 0.07 0.19*<0.0001 0.24 <0.0001**<0.0001 0.28 <0.0001<0.0001 0.20 <0.0001<0.0001 0.18 0.001<0.0001 0.20 <0.00010.07 0.02 0.580.001 �0.13 0.01<0.0001 0.17 0.0010.007 0.06 0.30<0.0001 0.18 <0.00010.001 0.14 0.0080.01 0.12 0.040.001 �0.11 0.04

<0.0001 �0.27 <0.0001

*P values refer to results after analyses with adjustment for age. **P values refer to

Table 3Univariate correlations between IMT and anthropometric and biochemical variables stratified by gender.

Unadjustedcorrelationsbetween IMT andmetabolic variables

P Age-adjustedcorrelationsbetween IMT andmetabolic variables

P Unadjustedcorrelationsbetween IMT andmetabolic variables

P Age-adjustedcorrelationsbetween IMT andmetabolic variables

P

Pearson’s correlationcoefficient (r)

Pearson’s correlationcoefficient (r)

Pearson’s correlationcoefficient (r)

Pearson’s correlationcoefficient (r)

Men Women

Age (yrs) 0.24 0.001 e 0.19 0.008 e

BMI (kg/m2) 0.17 0.02 0.16 0.03 0.32 <0.0001 0.29 <0.0001Waist circumference

(cm)0.12 0.09 0.10 0.18 0.22 0.001 0.18 0.01

SBP (mmHg) 0.19 0.01 0.10 0.10 0.24 0.001 0.17 0.01DBP (mmHg) 0.27 <0.0001 0.22 0.004 0.19 0.009 0.14 0.05Total cholesterol (mg/dl) 0.05 0.52 0.02 0.58 0.09 0.19 0.02 0.80HDL cholesterol (mg/dl) �0.03 0.77 �0.01 0.95 �0.18 0.01 �0.17 0.01Triglycerides (mg/dl) 0.13 0.08 0.08 0.27 0.28 <0.0001 0.22 0.002Fasting glucose (mg/dl) 0.14 0.05 0.10 0.19 0.06 0.41 0.03 0.722-h Glucose (mg/dl) 0.14 0.05 0.11 0.14 0.24 <0.0001 0.19 0.008Fasting insulin (mU/ml) 0.04 0.64 0.03 0.65 0.14 0.02 0.10 0.072-h Insulin (mU/ml) 0.09 0.22 0.07 0.33 0.18 0.01 0.15 0.02Insulin-stimulated

glucose disposal(mg � kg�1 � min�1)

�0.11 0.05 �0.12 0.05 �0.22 <0.0001 �0.23 <0.0001

Insulin clearance(ml/min � m2)

�0.11 0.05 �0.14 0.03 �0.36 <0.0001 �0.38 <0.0001

BMI ¼ body mass index; SBP ¼ systolic blood pressure; DBP ¼ diastolic blood pressure.

M.A. Marini et al. / Atherosclerosis 229 (2013) 453e458456

4. Conclusions

Insulin clearance is an integrated component of insulin meta-bolism and a strong determinant of the variability in both fasting[14] and postprandial insulin levels [17,18]. Reduced insulin clear-ance has been associated with conditions of insulin resistance suchas obesity [16e19] and impaired glucose tolerance [15], and it is anindependent predictor of type 2 diabetes [20]. Insulin clearancemay contribute to modulate cellular response to the hormone intarget tissues of insulin action including vasculature by regulatinginsulin availability, and insulin receptor levels [27]. As a majordeterminant of plasma insulin levels, insulin clearance may play arole in modulating processes that influence the development ofvascular atherosclerosis. In the present study, we investigated theassociation between insulin clearance and carotid IMT, a validatedmeasure of vascular damage [21,26], in a Caucasian cohort of non-diabetic offspring of type 2 diabetic patients. We found that insulinclearance was independently associated with carotid IMT. This as-sociation was stronger in women than in men. A multivariateregression analysis in a model including well-known cardio-metabolic risk factors showed that the only variables that remainedsignificantly associated with carotid IMT were BMI, insulin clear-ance, age, waist circumference, diastolic blood pressure, and 2-hpost-load glucose accounting for 26% of the variation in carotidIMT. Insulin clearance and BMI were the traits most stronglyassociated with carotid IMT. Similar results were observed whenmultivariate regression analysis was performed in the sub-group ofNGT (n ¼ 270) with the exception that 2-h post-load glucose wasreplaced by fasting plasma glucose as an independent variableassociated with carotid IMT. Lower insulin clearance values wereassociated with a 4.06-fold higher odds of having vascular damageas compared with those with higher values of insulin clearanceafter adjustment for several cardio-metabolic risk factors. Thus,reduction in insulin clearance appears to be independently asso-ciated with vascular damage in adult non-diabetic subjects. To thebest of our knowledge, this is the first study reporting the inde-pendent effect of insulin clearance on carotid IMT even aftercorrection for several confounders.

It has been suggested that in condition of insulin resistance suchas obesity, decreased insulin clearance may represent a compen-satory mechanism to alleviate pancreatic b-cell stress from exces-sive demand andmaintain normoglycemia [28]. On the other hand,sustained hyperinsulinemia due to decreased insulin clearancemayhave adverse effects on target tissues of insulin action includingvasculature. There is evidence that exposure of vascular cells toelevated concentrations of insulin results in proatherogenic effects.Indeed, in cultured endothelial cells elevated concentrations ofinsulin promote the secretion of endothelin-1 (ET-1), a potentvasoconstrictor protein [29], the expression of vascular cell adhe-sion molecule (VCAM)-1 and monocytes adhesion to culturedendothelial cells [30,31]. In addition, it has been demonstrated thathigh concentrations of insulin increase the expression of plasmin-ogen activator inhibitor (PAI-1) in human vascular smooth musclecells [32]. Interestingly, it has been reported a significant associa-tion between PAI-1 and insulin clearance in the Insulin ResistanceAtherosclerosis Study (IRAS) [33] thus supporting the idea thatchronic exposure to elevated concentrations of insulin may haveproatherogenic effects.

Nonetheless, the observation that insulin clearance was an in-dependent correlate of carotid IMT in a model that included fastingand 2-h post-load insulin argues against the possibility thathyperinsulinemia may mediate the adverse effects of insulinclearance on the vasculature.

Receptor-mediated insulin endocytosis is the principal mecha-nism for insulin clearance from the circulation [34]. There is evi-dence suggesting that internalization and intracellular processingof the insulinereceptor complex may play a role in insulin action[35,36]. Since it has been reported that insulin-resistant subjectshave primary defects in insulineinsulin receptor complex pro-cessing [37], it is possible that alteration in the internalization andprocessing of insulin in vascular cells may impairs insulin action invascular cells such as endothelial cells and smooth muscle cells,thus explaining, at least in part, the independent association be-tween insulin clearance and carotid IMT.

Strengths of present study include the inclusion of both sexes,the exclusion of confounding conditions characterized by elevation

M.A. Marini et al. / Atherosclerosis 229 (2013) 453e458 457

in carotid IMT such as subjects with history of cardiovascular dis-ease, the use of OGTT instead of fasting glucose alone to excludediabetic subjects, the homogeneity of the sample with detailedclinical characterization performed by a trained staff following astandardized protocol, the use of the gold standard hyper-insulinemic euglycemic clamp for assessment of insulin sensitivityand clearance, and the centralization of laboratory analyses.Nevertheless, some limitations should be considered in the inter-pretation of the present results. First, the cross-sectional design ofthe study precludes us to draw any conclusion on the role ofimpairment in insulin clearance in development of cardiovasculardisease, and, therefore, no conclusion regarding causeeeffect re-lationships can be made. It is in fact possible that increased IMTreflects changes in the vasculature that impair insulin clearance byimpeding the exit of insulin from the vascular space. Second, allbiochemical variables including OGTT were measured once.Although such an approach is a common limitation to most largeepidemiological studies, intra-individual variation cannot be takeninto account, and some subjects might have been misclassified.Moreover, the present results might have been influenced by thepresence of a family history of type 2 diabetes. However, type 2diabetes and cardiovascular disease share common genetic ante-cedents, and many individuals who develop cardiovascular diseasehave a family history of diabetes. Multivariate analysis included allsubjects, although there were minor differences in insulin sensi-tivity, and cardio-vascular conventional risk factors, between malesand females. However, gender differences were adjusted for in themultivariate models. Additionally, the factors that we consideredtogether explained 26% of the overall variation in carotid IMT,suggesting the existence of other regulators of vascular thickness.Finally, our findings are only based on Caucasian individuals, andresults might vary as a function of ethnic group. Caution in gener-alizing these results is warranted, as Caucasian individuals havebeen found to have higher insulin clearance than Mexican Ameri-cans [38], and Asian Indians [39].

Conflict of interest

All Authors declare that they have no conflict of interest.

References

[1] Bonora E, Kiechl S, Willeit J, Oberhollenzer F, Egger G, Meigs JB, et al. Popu-lation-based incidence rates and risk factors for type 2 diabetes in white in-dividuals: the Bruneck Study. Diabetes 2004;53:1782e9.

[2] Weyer C, Bogardus C, Mott DM, Pratley RE. The natural history of insulinsecretory dysfunction and insulin resistance in the pathogenesis of type 2diabetes mellitus. J Clin Invest 1999;104:787e94.

[3] Howard G, O’Leary DH, Zaccaro D, Haffner S, Rewers M, Hamman R, et al.Insulin sensitivity and atherosclerosis. The Insulin Resistance AtherosclerosisStudy (IRAS) Investigators. Circulation 1996;93:1809e17.

[4] Rutter MK, Meigs JB, Sullivan LM, D’Agostino Sr RB, Wilson PW. Insulinresistance, the metabolic syndrome, and incident cardiovascular events in theFramingham Offspring Study. Diabetes 2005;54:3252e7.

[5] Reaven G. Role of insulin resistance in human disease. Diabetes 1988;37:1595e607.

[6] Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, et al.Harmonizing the metabolic syndrome: a joint interim statement of theInternational Diabetes Federation Task Force on Epidemiology and Preven-tion; National Heart, Lung, and Blood Institute; American Heart Association;World Heart Federation; International Atherosclerosis Society; and Interna-tional Association for the Study of Obesity. Circulation 2009;120:1640e5.

[7] Despres JP, Lamarche B, Mauriege P, Cantin B, Dagenais GR, Moorjani S, et al.Hyperinsulinemia as an independent risk factor for ischemic heart disease.N Engl J Med 1996;334:952e7.

[8] Folsom AR, Szklo M, Stevens J, Liao F, Smith R, Eckfeldt JH. A prospective studyof coronary heart disease in relation to fasting insulin, glucose, and diabetes:the Atherosclerosis Risk in Communities (ARIC) Study. Diabetes Care 1997;20:935e42.

[9] Hu G, Qiao Q, Tuomilehto J, Eliasson M, Feskens EJ, Pyorala K., DECODE InsulinStudy Group. Plasma insulin and cardiovascular mortality in non-diabetic

European men and women: a meta-analysis of data from eleven prospectivestudies. Diabetologia 2004;47:1245e56.

[10] Pyorala M, Miettinen H, Laakso M, Pyorala K. Hyperinsulinemia predictscoronary heart disease risk in healthy middle-aged men: the 22-year follow-up results of the Helsinki Policemen Study. Circulation 1998;98:398e404.

[11] Ferrara A, Barrett-Connor EL, Edelstein SL. Hyperinsulinemia does not increasethe risk of fatal cardiovasculardisease inelderlymenorwomenwithoutdiabetes:the Rancho Bernardo Study, 1984e1991. Am J Epidemiol 1994;140:857e69.

[12] Lindberg O, Tilvis RS, Stranberg TE, Valvanne J, Sairanen S, Ehnholm C, et al.Elevated fasting plasma insulin in a general aged population: an innocentcompanion of cardiovascular diseases. J Am Geriatr Soc 1997;45:407e12.

[13] Lindahl B, Dinesen B, Eliasson M, Røder M, Jansson JH, Huhtasaari F, et al. Highproinsulin concentration precedes acute myocardial infarction in a nondia-betic population. Metabolism 1999;48:1197e202.

[14] Goodarzi MO, Cui J, Chen YD, Hsueh WA, Guo X, Rotter JI. Fasting insulinreflects heterogeneous physiological processes: role of insulin clearance. AmJ Physiol Endocrinol Metab 2011;301:E402e8.

[15] Bonora E, Zavaroni I, Coscelli C, Butturini U. Decreased hepatic insulinextraction in subjects with mild glucose intolerance. Metabolism 1983;32:438e46.

[16] Faber OK, Christensen K, Kehlet H, Madsbad S, Binder C. Decreased insulinremoval contributes to hyperinsulinemia in obesity. J Clin Endocrinol Metab1981;53:618e21.

[17] Bonora E, Zavaroni I, Bruschi F, Alpi O, Pezzarossa A, Guerra L, et al. Peripheralhyperinsulinemia of simple obesity: pancreatic hypersecretion or impairedinsulin metabolism? J Clin Endocrinol Metab 1984;59:1121e7.

[18] Erdmann J, Mayr M, Oppel U, Sypchenko O, Wagenpfeil S, Schusdziarra V.Weight-dependent differential contribution of insulin secretion and clearanceto hyperinsulinemia of obesity. Regul Pept 2009;152:1e7.

[19] Jones CN, Abbasi F, Carantoni M, Polonsky KS, Reaven GM. Roles of insulinresistance and obesity in regulation of plasma insulin concentrations. AmJ Physiol Endocrinol Metab 2000;278:E501e8.

[20] Lee CC, Haffner SM, Wagenknecht LE, Lorenzo C, Norris JM, Bergman RN, et al.Insulin clearance and the incidence of type 2 diabetes in Hispanics and AfricanAmericans: the IRAS Family Study. Diabetes Care 2013;36:901e7.

[21] O’Leary DH, Polak JF, Kronmal RA, Manolio TA, Burke GL, Wolfson Jr SK.,Cardiovascular Health Study Collaborative Research Group. Carotid arteryintima and media thickness as a risk factor for myocardial infarction andstroke in older adults. N Engl J Med 1999;340:14e22.

[22] Laakso M, Zilinskaite J, Hansen T, Boesgaard TW, Vänttinen M, Stancáková A,et al. Insulin sensitivity, insulin release and GLP-1 levels in subjects with IFGand/or IGT in the EUGENE2 study. Diabetologia 2008;51:502e11.

[23] Marini MA, Frontoni S, Mineo D, Bracaglia D, Cardellini M, De Nicolais P, et al.The Arg972variant in insulin receptor substrate-1 is associated with anatherogenic profile in offspring of type 2 diabetic patients. J Clin EndocrinolMetab 2003;88:3368e71.

[24] Cardellini M, Marini MA, Frontoni S, Hribal ML, Andreozzi F, Perticone F, et al.Carotid artery intima-media thickness is associated with insulin-mediatedglucose disposal in nondiabetic normotensive offspring of type 2 diabeticpatients. Am J Physiol Endocrinol Metab 2007;292:E347e52.

[25] DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method forquantifying insulin secretion and resistance. Am J Physiol Endocrinol MetabGastrointest Physiol 1979;237:E214e23.

[26] Mancia G, De Backer G, Dominiczak A, Cifkova R, Fagard R, Germano G, et al.2007 Guidelines for the management of arterial hypertension: the Task Forcefor the Management of Arterial Hypertension of the European Society ofHypertension (ESH) and of the European Society of Cardiology (ESC). EurHeart J 2007;28:1462e536.

[27] Abdul-Hay SO, Kang D, McBride M, Li L, Zhao J, Leissring MA. Deletion ofinsulin-degrading enzyme elicits antipodal, age-dependent effects on glucoseand insulin tolerance. PLoS One 2011;6:e20818.

[28] Mittelman SD, Van Citters GW, Kim SP, Davis DA, Dea MK, Hamilton-Wessler M, et al. Longitudinal compensation for fat-induced insulin resistanceincludes reduced insulin clearance and enhanced beta-cell response. Diabetes2000;49:2116e25.

[29] Potenza MA, Marasciulo FL, Chieppa DM, Brigiani GS, Formoso G, Quon MJ,et al. Insulin resistance in spontaneously hypertensive rats is associated withendothelial dysfunction characterized by imbalance between NO and ET-1production. Am J Physiol 2005;289:H813e22.

[30] Madonna R, Pandolfi A, Massaro M, Consoli A, De Caterina R. Insulin enhancesvascular cell adhesion molecule-1 expression in human cultured endothelialcells through a pro-atherogenic pathway mediated by p38 mitogen activatedprotein-kinase. Diabetologia 2004;47:532e6.

[31] Madonna R, De Caterina R. Prolonged exposure to high insulin impairs theendothelial PI3-kinase/Akt/nitric oxide signalling. Thromb Haemost2009;101:345e50.

[32] Pandolfi A, Iacoviello L, Capani F, Vitacolonna E, Donati MB, Consoli A. Glucoseand insulin independently reduce the fibrinolytic potential of human vascularsmooth muscle cells in culture. Diabetologia 1996;39:1425e31.

[33] Lee CC, Lorenzo C, Haffner SM, Wagenknecht LE, Festa A, Goodarzi MO, et al.The association of inflammatory and fibrinolytic proteins with 5 year changein insulin clearance: the Insulin Resistance Atherosclerosis Study (IRAS).Diabetologia 2013;56:112e20.

[34] Duckworth WC, Bennett RG, Hamel FG. Insulin degradation: progress andpotential. Endocr Rev 1998;19:608e24.

M.A. Marini et al. / Atherosclerosis 229 (2013) 453e458458

[35] DuckworthWC, Bennett RG, Hamel FG. The significance of intracellular insulinto insulin action. J Investig Med 1997;45:20e7.

[36] Hribal ML, D’Alfonso R, Giovannone B, Lauro D, Liu YY, Borboni P, et al. Thesulfonylurea glimepiride regulates intracellular routing of the insulin-receptorcomplexes through their interaction with specific protein kinase C isoforms.Mol Pharmacol 2001;59:322e30.

[37] Sesti G, D’Alfonso R, Vargas Punti MD, Tullio AN, Liu YY, Federici M, et al.Delayed intracellular dissociation of the insulin-receptor compleximpairs receptor recycling and insulin processing in cultured EBV-

transformed lymphocytes from insulin-resistant subjects. Diabetologia1996;39:289e97.

[38] Haffner SM, Stern MP, Watanabe RM, Bergman RN. Relationship of insulinclearance and secretion to insulin sensitivity in non-diabetic Mexican Amer-icans. Eur J Clin Invest 1992;22:147e53.

[39] Liew CF, Seah ES, Yeo KP, Lee KO, Wise SD. Lean, nondiabetic Asian Indianshave decreased insulin sensitivity and insulin clearance, and raised leptincompared to Caucasians and Chinese subjects. Int J Obes Relat Metab Disord2003;27:784e9.