Carotid Artery Wall Intima-Media Thickness Is Associated With Insulin-Mediated Glucose Disposal in Men at High and Low Coronary Risk
Background and Purpose The aim of this investigation was to examine the relationship between insulin sensitivity and intima-media thickness in the common carotid artery.
Methods Ultrasound examinations of the common carotid artery and hyperinsulinemic euglycemic clamp examinations were performed in a group (n=25) of men aged 57 to 77 years at high risk for atherosclerotic disease (hypertension and at least one of the following factors: hypercholesterolemia and/or smoking) and in an age-matched low-risk group (n=23) with no cardiovascular risk factors. Subjects with cardiovascular disease or diabetes mellitus were excluded.
Results A significant negative relationship between insulin sensitivity index and common carotid maximum intima-media thickness was observed in both the high-risk group (r=−.45, P<.05) and in the low-risk group (r=−.59, P<.01).
Conclusions Our results suggest that an increase in intima-media thickness, as a possible expression of early atherosclerosis, is negatively related to insulin sensitivity.
It has been suggested that insulin resistance may play a pathophysiological role in the atherosclerotic process.1 The postulation of such a mechanism relates to several observations. First, insulin resistance is associated with a number of known cardiovascular risk factors: hypertension,2 3 diabetes mellitus,1 dyslipidemia,4 smoking,5 6 hemostatic and fibrinolytic factors,7 microalbuminuria,8 9 and upper body obesity.10 Second, insulin resistance is in general accompanied by hyperinsulinemia,11 and it has been proposed that circulating insulin may contribute to the atherogenic process. Several observations support this hypothesis: Prospective studies indicate that hyperinsulinemia is a risk factor for cardiovascular disease,12 13 and it has been demonstrated that insulin has trophic effects on tissue components in the atherosclerotic lesion.1 14 15 16
Noninvasive techniques such as B-mode ultrasound can more directly assess intima-media thickness, thereby providing an opportunity to study early phases in the atherosclerotic disease process. In the Atherosclerosis Risk in Communities (ARIC) study,17 abdominal adiposity, abnormal glucose metabolism, and fasting plasma insulin were associated positively with mean intima-media thickness of the carotid artery in subjects without cardiovascular disease. These results give support to the hypothesis that insulin resistance may be one factor of pathogenetic importance in the development of cardiovascular disease.
We had an opportunity to address this issue within the framework of an ongoing study of a group of hypertensive men and a healthy reference group. Thus, we examined whether insulin-mediated glucose uptake is related to intima-media thickness in the carotid artery in asymptomatic men.
Subjects and Methods
Five hundred eight male patients with treated hypertension were originally included in a risk factor intervention study.18 The inclusion criteria of this intervention study, apart from treated hypertension and male sex, were age between 50 and 72 years and one or more of the following: hypercholesterolemia (serum cholesterol ≥6.5 mmol/L), tobacco smoking (one or more cigarettes per day), or diabetes mellitus.19 The diagnosis of primary hypertension had been established according to previously reported routines.18
The background population was representative of high-risk hypertensive subjects in Göteborg, Sweden, since the majority was earlier recruited by screening a random third of all men in their respective age groups in Göteborg.20 In the aforementioned intervention study, the patients had been randomized either to a multiple risk factor treatment program or to conventional treatment. The former was based on a nonpharmacological and if necessary a pharmacological regimen that aimed to lower hypercholesterolemia and also on a smoking cessation program.18
From this group of 508 men, one third of the patients were randomly selected to take part in an ultrasound study of the carotid region. Of 169 patients randomized to the ultrasound study, 164 patients agreed to take part. These patients have previously been described.21 After 3 years of follow-up, 141 of the patients were reexamined with B-mode ultrasound of the carotid artery. From this group of patients examined by ultrasound, 40 nondiabetic patients were randomly selected and accepted to participate. Exclusion criteria in the present study were, in accordance with the ARIC study, previous or current cardiovascular disease as defined below. We also chose to exclude patients with diabetes mellitus19 because such patients are characterized by insulin resistance.11 Twenty-five patients fulfilled the criteria and were included in the analyses.
A sex- and age-matched randomly selected population sample of 670 men living in Göteborg were sent a letter inviting those who felt healthy and had been nonsmokers for at least 3 years to take part in the study. Subjects who in writing expressed their interest to take part in the study were offered a screening examination (n=142).
These volunteers were included in the low-risk group if they fulfilled the following inclusion criteria: diastolic blood pressure <95 mm Hg, no antihypertensive treatment, no smoking during the last 3 years, serum cholesterol ≥6.5 mmol/L, normal fasting blood glucose,19 and sinus rhythm on an electrocardiographic (ECG) examination. A total of 53 subjects were included in the low-risk group. Twenty-five of these subjects were randomly selected for clamp examinations, and 23 fulfilled the inclusion criteria and were included in this study.
All subjects gave informed consent after they received written and oral information, and the study was approved by the ethics committee of the Faculty of Medicine, Göteborg University.
Resting blood pressure was measured phonographically (Korotkoff sounds recorded on ECG paper) in the patient’s right arm after rest in the supine position in connection with the ultrasound examination as described earlier.22 Blood pressure was calculated to the nearest 1 mm Hg, and the mean of two recordings was used. Body weight, body mass index, and the ratio of waist to hip circumference were measured according to recommended principles.22 Smoking was assessed by a questionnaire. The total number of years of smoking was multiplied by the average number of cigarettes smoked daily; the product was termed “cigarette-years.” Manifest cardiovascular disease was defined as either history of cardiovascular disease (one or more of the following diagnoses: stroke, transient ischemic attack, myocardial infarction, angina pectoris, intermittent claudication) or a major finding according to the Minnesota code23 : definite Q or QS wave (1:1 or 1:2), definite ST or T wave (4:1-2, 5:1-2), left bundle branch block (7:1), or a wide QRS complex >0.12 second (7:4). Established criteria for stroke, transient ischemic attack, myocardial infarction, angina pectoris, and intermittent claudication were used.18
Venous blood was drawn after an overnight fast and after 5 minutes of supine rest for determination of blood glucose, serum levels of triglycerides, and total and high-density lipoprotein cholesterol with the use of established methods.18
Hyperinsulinemic Euglycemic Clamp
The hyperinsulinemic euglycemic clamp examinations were performed as previously described.24 After the patients fasted overnight, an indwelling catheter was inserted into a brachial vein for glucose and insulin infusion. A second catheter was placed in an antecubital vein in the contralateral arm. The arm was warmed with heating pads to arterialize the blood. After a 10-minute infusion of a priming dose, insulin (Actrapid, Novo-Nordisk), at a concentration of 0.5 IU/mL dissolved in isotonic saline, was infused at a constant rate with an infusion pump (IMED 922 H). The insulin was given at the infusion rate of 1.0 mU/kg body wt per minute. Blood glucose levels were kept constant by the continuous venous infusion of glucose (200 mg/mL). Blood for plasma glucose levels was drawn every 5 minutes. Blood was also obtained for later determination of plasma insulin at 60 and 120 minutes after the start of insulin infusion. When steady state had been reached, which required approximately 60 minutes, the glucose disposal rate was calculated during the last 60 minutes of the total insulin infusion period, ie, between 60 and 120 minutes. Under steady-state conditions the rate of glucose infusion is equal to the rate of glucose disposal, provided that endogenous hepatic glucose production is absent. The insulin levels (mean, 69 mU/L) obtained during the insulin infusion have been reported to suppress hepatic glucose production to a negligible rate even in hypertensive subjects with the same body mass index (mean, 26 kg/m2), as in this study.3 The serum concentration of C peptides was also examined to verify that the endogenic pancreatic secretion of insulin was inhibited during the clamp investigation, which it was in all subjects (data not shown). The glucose disposal during the clamp was expressed as the amount of glucose infused per kilogram lean body weight per minute during the last 60 minutes of the clamp examination. The insulin sensitivity index was calculated as glucose disposal per kilogram body weight divided by the mean plasma insulin concentration multiplied by 100.2
Total body potassium was determined in a whole-body counter that detected naturally occurring 40K (Nuclear Enterprise Ltd). Lean body mass was estimated according to Forbes et al,25 assuming that 1 kg lean body mass equals 68.1 mmol potassium. Body fat was calculated by subtracting the lean body mass from the body weight. Lean body mass was obtained in 23 high-risk and 22 low-risk subjects.
Subjects were examined in a supine position with an ultrasound scanner (Acuson 128) equipped with a linear 7-MHz transducer, as earlier described in detail.21 26 The distal part of the right common carotid artery was scanned by modifying the ultrasound beam to pass perpendicular to the vessel wall to achieve the typical two-line image of the vessel wall structures from both the anterior and posterior walls. At the position of the best visibility of the far wall intima-media complex, three images were captured by ECG triggering21 26 and recorded on videotape.
The video-recorded frozen images were analyzed off-line in a computerized analyzing system along a 10-mm-long section just proximal to the carotid bulb.21 26 Intima-media thickness was defined as the distance from the leading edge of the lumen-intima interface of the far wall to the leading edge of the media-adventitia interface of the far wall. The computer program calculated the maximum and mean values of intima-media thickness.
Interobserver variability (including variation in data collection and measurements) studied in our laboratory with repeated recordings by two independent observers has shown a coefficient of variation for mean intima-media thickness of 10.2% and for maximum intima-media thickness of 8.9%.
Plaques in the Carotid Artery. To identify and record the occurrence of atherosclerotic plaque, the carotid artery was scanned from the distal part of the common carotid artery and approximately 10 mm further in the external and internal carotid arteries.21 26
At the position of the best visibility of a plaque, ie, the largest cross-sectional area in a longitudinal transaxial view, sometimes attained by guidance of successive cross-sectional views along the plaque extension, three images were captured (triggered by R waves).
A semiquantitative subjective scale (visual scoring) was used to grade the size of plaques in the four locations in the carotid artery region: external and internal carotid artery, carotid bulb, and distal part of the common carotid artery. This analysis included plaques in the near and far walls of the vessel.
A plaque was defined as a distinct area with an intima-media thickness >50% thicker compared with neighboring sites judged visually,21 26 as follows: grade 0, no plaque; grade 1, ≥1 small plaque (each <≈10 mm2); grade 2, moderately sized plaques (the differentiation between grades 1 and 2 was made subjectively in most cases, and quantitative measurements of the area were made in the computerized analyzing system21 26 only when the correct classification was not obvious to the observer); grade 3, large plaques that cause a change in blood flow defined by the pulsed Doppler curve: peak systolic velocity >1.2 m/s and ≤60° Doppler angle.21 26
Results are presented as means and SDs. Continuous variables were compared with the Mann-Whitney U test. The measure of insulin resistance, the insulin sensitivity index, was based on the ratios between glucose infusion rate, time, body weight, and plasma insulin concentration. Because this variable cannot be expected to be normally distributed, a nonparametric method (Spearman rank correlation coefficient) was used in the correlation analysis. A nonparametric method of multivariate analysis was applied for testing the correlation between two variables when the influence of a third variable was eliminated by use of Mantel’s test. A two-sided P<.05 was considered statistically significant.
Characteristics of the high-risk and low-risk groups are shown in Table 1⇓. The high-risk group had higher serum triglyceride concentration and higher waist-hip ratio compared with the low-risk group. In the high-risk group all patients were on antihypertensive treatment, and 16% were on lipid-lowering drugs.
In the high- and low-risk groups there was a significant negative relationship between insulin sensitivity index and maximum common carotid intima-media thickness (r=−.45, P=.027 and r=−.59, P=.006, respectively; Table 2⇓, Figure⇓). A significant negative relationship between maximum intima-media thickness and glucose disposal adjusted for lean body mass was observed in the low-risk group (Table 2⇓).
Body mass index showed a positive significant relationship with carotid maximum intima-media thickness in the low-risk group (Table 2⇑). No other significant correlations with intima-media thickness were found in any of the groups.
In a multivariate analysis in which Mantel’s test was used, the relationship between maximum carotid intima-media thickness and insulin sensitivity index in the low-risk group remained significant after adjustment for the confounding effect of body mass index (P=.007).
The plaque status was not associated with the insulin sensitivity index (data not shown).
The main result of our study was that the insulin-mediated glucose uptake was negatively associated with the maximum carotid intima-media thickness in both the asymptomatic hypertensive men at high cardiovascular risk and in the low-risk group, thereby supporting the concept of a relationship between peripheral insulin metabolism and carotid artery wall morphology.
This finding has to be considered from several methodological aspects. First, the results were obtained as secondary findings in an ongoing study. The limited sample size raises the possibility of a statistical type I error. However, this is highly improbable because similar findings were obtained in both patients and low-risk subjects.
Second, the observation that the high-risk hypertensive group tended to have a lower insulin sensitivity index, ie, they were more insulin resistant than the low-risk group, is, on the one hand, in line with current knowledge.2 3 On the other hand, not only hypertension but also smoking5 6 and treatment with some antihypertensive27 and lipid-lowering drugs28 are known to affect insulin sensitivity. We cannot exclude that current drug therapy in part may explain the results in the hypertensive group. However, the relationship between insulin sensitivity and intima-media thickness was also observed in the untreated low-risk group.
We used two established methods to adjust insulin-mediated glucose uptake for body size: the insulin sensitivity index, which takes body weight and current plasma insulin into account, and a method that adjusts glucose disposal for lean body mass. The latter method is preferable because it expresses glucose uptake in relation to muscle mass, which is the major determinant of glucose elimination. However, it was not possible to perform measurement of lean body mass in all patients. Hence, there were fewer patients to include in the analysis when this variable was used.
Our population samples were not representative of the general population. However, the selection processes are well characterized, and the two study groups may represent both individuals at high risk of cardiovascular disease and those at low risk.
Only measurements of the right common carotid artery were done, and they cannot be claimed to be an overall measurement of the atherosclerotic process. Previous studies have shown that different arterial regions seem to differ in relation to different established risk factors for atherosclerotic disease.29
The high- and low-risk men did not differ significantly in carotid artery intima-media thickness. The explanation is possibly that the high-risk men had been treated for a long time.
With reservations for the aspects discussed above, the observations in the present study are in accordance with the results from the ARIC study,17 which revealed an association between carotid artery intima-media thickness and body mass index, waist-hip ratio, physical inactivity, diabetes mellitus, and fasting insulin in a large population study of subjects without cardiovascular disease. All these variables have in common a consistent relationship to insulin sensitivity.2 3 4
A Finnish study has shown that in asymptomatic subjects examined with the ultrasound technique, those with plaque in the carotid and/or femoral artery were more insulin resistant than those without such findings.30 This result is in contrast with that in the present study, which shows no such association. There are several possible explanations for this discrepancy. Whereas our study dealt with hypertensive men at high coronary risk and a group of nonsmoking normotensive men without hypercholesterolemia, the Finnish study examined a highly selected, healthy group of subjects who were 15 years younger than our subjects. The designs of the two studies were also different. Finally, a type II error cannot be excluded.
In our study we excluded patients with diabetes mellitus, which further strengthens the concept of a relationship between insulin resistance and mechanisms promoting atherosclerosis. The covariation between intima-media thickness, insulin-mediated glucose disposal, and body mass index in the low-risk group may have several explanations. It can be argued that obesity might cause both an increase in arterial vessel wall thickness and a decrease in insulin sensitivity. This possibility seems to be refuted by the fact that after adjustment for body mass index there remained an independent, significant negative association between insulin sensitivity and maximum carotid intima-media thickness (P=.007) in the low-risk group. In addition, there was a significant association between glucose disposal adjusted for lean body mass and the maximum intima-media thickness in this group.
However, an increase in body mass index and a decrease in insulin sensitivity may well be the result of the same metabolic disorder, and it is therefore not self-evident that body mass index should be handled as a confounder in the analysis performed. Furthermore, in the high-risk group no association between maximum carotid intima-media thickness and body mass index was found, even though a significant negative relationship between maximum carotid intima-media thickness and insulin sensitivity was observed.
Insulin resistance may play a central role in the development of atherosclerosis both as a result of its relationship to different cardiovascular risk factors and also as an effect of suggested trophic mechanisms of insulin.1 Insulin can have many important effects, such as stimulating the growth of vascular smooth muscle cells,14 increasing the uptake of lipoprotein cholesterol in subintimal smooth muscle and fibroblast cells,15 decreasing the removal of cholesterol from foam cells in the subintimal portion of the vessel,1 and increasing the release of insulin-like growth factors.16
In conclusion, our results support the findings from a large population study17 and suggest a significant negative relationship between insulin sensitivity and maximum carotid intima-media thickness in both the low-risk and high-risk groups. This finding supports the concept that lowered insulin sensitivity may be associated with the atherosclerotic process.
This study was supported by grants from the Swedish Medical Research Council (B92-19X-09937-01A, B93-19X-09937-02B, B93-19X-03506, B94-19X-09937-03A), the Swedish Heart and Lung Foundation, the Göteborg Medical Society, and the Swedish Hypertension Society.
- Received January 3, 1995.
- Revision received February 13, 1995.
- Accepted March 6, 1995.
- Copyright © 1995 by American Heart Association
Laakso M, Sarlund H, Mykkänen L. Insulin resistance is associated with lipid and lipoprotein abnormalities in subjects with varying degrees of glucose tolerance. Arteriosclerosis. 1990;10:223-231.
DeFronzo RA, Ferrannini E. Insulin resistance: a multifaceted syndrome responsible for NIDDM, obesity, hypertension, dyslipidemia, and atherosclerotic cardiovascular disease. Diabetes Care. 1991;14:173-194.
Pyörälä K. Relationship of glucose tolerance and plasma insulin to the incidence of coronary heart disease: results from two population studies in Finland. Diabetes Care. 1979;2:131-141.
Folsom AR, Eckfeldt JH, Weitzman S, Ma J, Chambless LE, Barnes RW, Cram KB, Hutchinson RG, for the Atherosclerosis Risk in Communities (ARIC) Study Investigators. Relation of carotid artery wall thickness to diabetes mellitus, fasting glucose and insulin, body size, and physical activity. Stroke. 1994;25:66-73.
World Health Organisation Study Group. Diabetes Mellitus. Geneva, Switzerland: WHO Technical Report Series; 1985. No. 727.
Wilhelmsen L, Berglund G, Elmfeldt D, Tibblin G, Wedel H, Pennert K, Vedin A, Wilhelmsson C, Werkö L. The multifactor primary prevention trial in Göteborg, Sweden. Eur Heart J. 1986;7:279-288.
Suurküla M, Agewall S, Fagerberg B, Wendelhag I, Widgren B, Wikstrand J. Ultrasound evaluation of atherosclerotic manifestations in the carotid artery in high-risk hypertensive patients. Arterioscler Thromb. 1994;14:1297-1304.
Beckman Suurküla M, Wikstrand J, Berglund G, Sivertsson R. Body weight is more important than family history of hypertension for left ventricular function. Hypertension. 1991;17:661-668.
Rose GA, Blackburn H. Cardiovascular Survey Methods. Geneva, Switzerland: WHO Monograph Series; 1968. No. 56.
DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol. 1979;237:214-233.
Forbes GB, Gallup J, Hursh JB. Estimation of total body fat from potassium-40 content. Science. 1961;133:101-102.
Wendelhag I, Wiklund O, Wikstrand J. Arterial wall thickness in familial hypercholesterolemia: ultrasound measurement of intima-media thickness in the common carotid artery. Arterioscler Thromb. 1992;12:70-77.
Lithell H. Effect of antihypertensive drugs on insulin, glucose and lipid metabolism. Diabetes Care. 1991;14:203-209.
Alberti KGMM, Jones IR, Laker MF, Swai ABM, Taylor R. Effect of bezafibrate on metabolic profiles in non-insulin-dependent diabetes mellitus. J Cardiovasc Pharmacol. 1990;16(suppl 9):S21-S25.
Salonen JT, Salonen R. Ultrasound B-mode imaging in observational studies of atherosclerotic progression. Circulation. 1993;87(suppl II):II-56-II-65.
Laakso M, Sarlund H, Salonen R, Suhonen M, Pyörälä K, Salonen JT, Karhapää P. Asymptomatic atherosclerosis and insulin resistance. Arterioscler Thromb. 1991;11:1068-1076.