The clinical significance of carotid intima-media thickness in cardiovascular diseases: a survey in Beijing

Transcription

1 (2008) 22, & 2008 Nature Publishing Group All rights reserved /08 $ ORIGINAL ARTICLE The clinical significance of carotid intima-media thickness in cardiovascular diseases: a survey in Beijing L Liu 1, F Zhao 1, Y Yang 1,LTQi 1, BW Zhang 1, F Chen 1, D Ciren 1, B Zheng 1, SY Wang 2, Y Huo 1 and LS Liu 2 1 Division of Cardiology, Peking University First Hospital, Beijng, China and 2 Beijing Hypertension League Institute, Beijng, China This study is to investigate the relationships between prevalent cardiovascular events (myocardial infarction, stroke and peripheral arterial disease) and carotid intima-media thickness (CIMT) in middle-aged and older adults; to assess which of the indexes, IMT in the common carotid artery (CCA), internal carotid artery (ICA) and carotid bifurcation (CB) separately or in combination, is a better correlate. IMT of the CCA, ICA and CB were measured with duplex ultrasound in 1058 individuals (aged between 37 and 86 years old) in a suburban community of Beijing. Histories of cardiovascular events as well as prevalent risk factors were obtained. CIMT were compared between groups with and without cardiovascular events. Associations of cardiovascular events with IMT measurements in CCA, CB and ICA were ascertained. The group with cardiovascular events had higher mean CIMT (0.74 ( ) mm vs 0.65 ( ) mm, P ¼ 0.000) and higher abnormal CIMT proportion (77.20 vs 64.45%, P ¼ 0.000). The odds ratio (OR) for myocardial infarction was 1.74 (P ¼ 0.010), for stroke 1.98 (P ¼ 0.001) and for peripheral arterial disease 1.80 (P ¼ 0.072) in abnormal CIMT. However, after adjustment of traditional risk factors, the correlations disappeared, implying that CIMT may act via other risk factors. After adjustment of age, mean CIMT correlated best with total cardiovascular events (OR: 4.39 ( ), P ¼ 0.000) and stroke (OR: 4.98 ( ), P ¼ 0.000) separately; mean posterior CIMT correlated with myocardial infarction best (OR: 2.97 ( ), P ¼ 0.000). CIMT may act as an intermediate point for cardiovascular diseases. Combined CIMT might be the best index associated with cardiovascular diseases. (2008) 22, ; doi: /sj.jhh ; published online 25 October 2007 Keywords: carotid intima-media thickness; cardiovascular events; epidemiological study Introduction Atherosclerosis is a progressive diffused arterial disease. Lesions in carotid and coronary arteries are well correlated. 1 The carotid artery is viewed as a window on the coronary arteries. The atherosclerotic lesions detected by carotid ultrasound, such as carotid wall stiffness, increased carotid intima-media thickness (CIMT), plaque, ulcer or narrow lumens, are powerful predictors of future cardiovascular events. 2 Among these indexes, CIMT has drawn more and more attention. As an important tool in evaluating sub-clinical atherosclerosis burden and atherosclerotic process, CIMT is closely associated with many traditional cardiovascular risk factors (such as cholesterol, Correspondence: Dr F Zhao, Division of Cardiology, Peking University First Hospital, No 1, Street Dahongluochang, District Xicheng, Beijing , China. zhaofbmu@vip.sina.com Received 9 May 2007; revised 24 September 2007; accepted 24 September 2007; published online 25 October 2007 diabetes, blood pressure and smoking), some new risk factors (such as lipoprotein, platelet aggregability and hyperhomocysteinaemia), and target organ damages (such as left ventricular hypertrophy, microalbuminuria and decreased ankle brachial index). Increased CIMT provides a non-invasive measure of the cumulative atherosclerotic burden experienced by the individual in response to all risk factors over the lifespan. 3 5 CIMT also has a strong association with future clinical outcome. Multiple prospective epidemiological studies have shown that an increase in CIMT is associated with an increase in relative risk for myocardial infarction (MI) and stroke, and a decrease in CIMT is associated with a decrease in the incidence of cardiovascular events. CIMT plays a role in predicting the risk of cardiovascular diseases and in turn provides information on relevant therapeutic strategies, making it a surrogate marker or an intermediate end point for observational studies and interventional therapy. 2,4 7 Although it is clear that CIMT is a good predictor of subsequent cardiovascular diseases, it is important to

2 260 establish whether the incremental information obtained by carotid ultrasound further defines an individual s risk profile, which is still in controversy 8 and data to indicate that measurement of CIMT may be useful in symptomatic patients to further distinguish patients into higher-risk or lower-risk group are presently not available. 5 Since measurement of arterial wall IMT made with B-ultrasound used in 1986, great developments had been achieved. However, there is no uniform standard as to the measurement of CIMT in clinical practice environments, which represents a barrier to routine use of CIMT. 5 CIMT is easy to measure on the far wall of the common carotid artery (CCA). 3 Different site-specific imaging protocols were used in various trials, making comparison of the results confusing. The establishment of a standardized method of measuring CIMT is critical for further utilization and in urgent need. 4,8 The purpose of this study is to investigate the relationships between cardiovascular events (MI, stroke and peripheral arterial disease (PAD)) and CIMT in middle-aged and older adults in Beijing, China; to assess which of the indexes, IMT in the CCA, internal carotid artery (ICA) and carotid bifurcation (CB) separately or in combination, is a better correlate. Methods Population The Atherosclerosis Survey enrolled the cohort in the community of the Capital Steel Corporation set up by Beijing Hypertension League Institute, including 1058 subjects (508 females), aged between 37 and 86 years old. The cross-sectional data collected in 2005 were used. The study was approved by the institutional medical ethics committee, and informed consent was obtained. Assessment of cardiovascular events and risk factors All cardiovascular events (including MI, stroke and PAD) and risk factor status (including hypertension and diabetes) were adjudicated based on a review of data collected from hospitalizations and outpatient records. MI is defined by a history of acute MI; and if pathological Q waves or coronary T waves are noted in the electrocardiogram (ECG) and correspondingly regional wall motion abnormality is confirmed by echocardiography, MI is also diagnosed although some people have not shown obvious symptoms of heart attack before. Stroke is diagnosed by a history of acute cerebral infarction, intracerebral haemorrhage due to cerebrovascular diseases (which are confirmed by a CT or MRI scan) or transient ischaemic attack. PAD is diagnosed when the subject suffers from intermittent claudication, the peripheral pulses are diminished and Doppler ultrasound shows plaques at the arterial walls and more than 50% luminal stenosis. Subjects were classified as cigarette smokers if they had reported cigarette smoking before no matter whether they quitted. Height, weight, waist and hip circumferences were measured and body mass index (BMI) and waist hip ratio (WHR) were calculated. Sitting blood pressure (after 15 min of rest) was measured at the right upper arm using a mercury sphygmomanometer three times at 5-min intervals. Mean systolic blood pressure (SBP) and diastolic blood pressure (DBP) were calculated, and heart rate (HR) was recorded. Hypertension was defined as an SBP of 140 mm Hg or over, or a DBP of 90 mm Hg or over or current anti-hypertensive drugs for known hypertension. A fasting blood sample was collected for analysis of the following parameters using standard techniques in Beijing Hypertension League Institute, including fasting blood glucose (FBG), total cholesterol (TCHO), triglyceride (TG) and high-density lipoprotein cholesterol (HDL). Part of subjects without known diabetes had oral glucose tolerance test (OGTT). Newly diagnosed diabetes mellitus was defined by fasting glucose X7 mmol l 1 and 2-h glucose in OGTT X11.1 mmol l 1. Previous and newly diagnosed diabetic patients were summed as diabetes. Measurement of CIMT Carotid ultrasonography was conducted by three experienced ultrasonologists blinded to clinical information with a high-resolution 7 10 MHz linear-array transducer (GE vivid 7, Milwaukee, WI, USA). Images were obtained at the level of the distal 1.0 cm of CCA, CB and the proximal 1.0 cm of ICA of the left and right carotid arteries. On a longitudinal, two-dimensional ultrasound image, the carotid walls are displayed as two bright white lines separated by a hypoechogenic space. The distance between the leading edge of the inner bright line of the wall (lumen intima interface) and the leading edge of the outer bright line (media adventitia interface) indicates the IMT. When an optimal image was obtained, it was frozen on the R wave of the ECG and stored on videotape. The actual measurements of CIMT were performed offline. For each segment (CCA, CB and ICA), CIMT at two sites (with about 0.5 cm between) of the anterior and posterior walls were measured separately. For each individual, mean CIMT (meanimt) was determined as the average of IMT values at the 24 sites examined. The averages of IMT values at eight sites of each segment (CCA, CB and ICA) examined were calculated and defined as mean CCA-IMT (meancca), mean CB-IMT (meancb) and mean ICA- IMT (meanica) respectively. The mean anterior CIMT (antimt) and mean posterior CIMT (postimt) were determined as the averages of IMT values at 12 sites of anterior and posterior walls separately. The mean anterior and posterior CIMT at each segment were calculated and defined as mean anterior CCA-IMT

3 (antcca), mean posterior CCA-IMT (postcca), mean anterior CB-IMT (antcb), mean posterior CB-IMT (postcb), mean anterior ICA-IMT (antica) and mean posterior ICA-IMT (postica) respectively. The maximum CIMT (maximt) was also recorded. According to the carotid ultrasonography, subjects with maximt o0.9 mm were defined as normal, while with maximt X0.9 mm abnormal. The abnormal group included subjects with wall thickening (0.9 mmpmaximt o1.2 mm) and with carotid plaque (localized thickenings protruding into the lumen with maximt X1.2 mm). A reproducibility study was conducted. The intraobserver variability was 5.51% (0.04 mm), and the inter-observer variability was 8.64% (0.063 mm). Statistical analyses Continuous variables with normal distribution, such as SBP, DBP and BMI, were expressed as mean7s.d., and independent t-test was used to compare between two groups. Continuous variables with obvious skew distribution, such as age, FBG, TG and CIMT, were expressed as median and quartiles, and Mann Whitney U-test was used to compare between two groups. Results were presented as percentages for categorized variables and Pearson s w 2 -test was applied. The associations between CIMT and cardiovascular events were evaluated by use of a multiple logistic regression model. Analyses were performed with CIMT both used as a categorized variable (normal or abnormal) and as a continuous variable. P-value o0.05 was considered significant. All reported P-values were two sided. All analyses were performed using SPSS 12.0 software for Windows. Results Table 1 describes general characteristics of the study subjects. The relationship between CIMT and cardiovascular events The subjects with prevalent cardiovascular events (including MI, stroke and PAD) have higher mean CIMT and higher abnormal CIMT proportion (Table 2). Abnormal CIMT rate is higher in the group with PAD, compared to the group without PAD (78.95 vs 67.53%). However, that does not show statistical significance (P ¼ 0.07), which might be due to the limited number in the group with PAD (n ¼ 57). Table 3 shows the correlations of CIMT with cardiovascular events. In the whole population, the group with abnormal CIMT has higher risk of cardiovascular events. Compared to the normal CIMT group, the odds ratio (OR) for total cardiovascular events is 1.87 (P ¼ 0.000), for MI 1.74 (P ¼ 0.010), for stroke 1.98 (P ¼ 0.001) and for PAD 1.80 (P ¼ 0.072) in the group with abnormal CIMT. In the male population, the OR for total cardiovascular events is 2.40 (P ¼ 0.000), for MI 2.60 (P ¼ 0.004), for stroke 2.41 (P ¼ 0.004) and for PAD 4.31 (P ¼ 0.032) in the group with abnormal CIMT. Such results are not found in female population. However, after adjustment of traditional cardiovascular risk factors (including age, blood pressure, lipids, diabetes, WHR and smoking and so on), the correlations disappear, implying that CIMT may act via other risk factors. 261 Table 1 Demographics and characteristics of the study population Characteristic All subjects (n ¼ 1058) Male (n ¼ 550) Female (n ¼ 508) Age (years) ( ) ( ) ( )** Hypertension (%) Diabetes mellitus (%) Smokers (%) ** Myocardial infarction (%) ** Stroke (%) PAD (%) Total events (%) ** BMI (kg m 2 ) WHR ** HR (b.p.m.) SBP (mm Hg) DBP (mm Hg) ** TCHO (mmol l 1 ) ** TG (mmol l 1 ) 1.82 ( ) 1.75 ( ) 1.94 ( )** HDL (mmol l 1 ) ** FBG (mmol l 1 ) 5.55 ( ) 5.52 ( ) 5.58 ( ) Mean IMT (mm) 0.68 ( ) 0.73 ( ) 0.61 ( )** Abnormal CIMT (%) ** Carotid plaque (%) ** Abbreviations: BMI, body mass index; CIMT, carotid intima-media thickness; DBP, diastolic blood pressure; FBG, fasting blood glucose; HDL, high-density lipoprotein; HR, heart rate; PAD, peripheral arterial disease; SBP, systolic blood pressure; TCHO, total cholesterol; TG, triglyceride; WHR, waist hip ratio.**po0.01, when compared to male subjects.

4 262 Table 2 Comparison of CIMT between groups with and without events MeanIMT (mm) Abnormal CIMT (%) Total events No (n ¼ 751) 0.65 ( ) Yes (n ¼ 307) 0.74 ( )** 77.20** MI No (n ¼ 923) 0.66 ( ) Yes (n ¼ 135) 0.75 ( )** 77.78* Stroke No (n ¼ 884) 0.66 ( ) Yes (n ¼ 174) 0.77 ( )** 79.31** PAD No (n ¼ 1001) 0.67 ( ) Yes (n ¼ 57) 0.71 ( )* Abbreviations: CIMT, carotid intima-media thickness; MI, myocardial infarction; PAD, peripheral arterial disease. *Po0.05, **Po0.01, when compared to subjects without events. Table 3 The correlations of CIMT with cardiovascular events Unadjusted Fully adjusted OR (95% CI) P OR (95% CI) P Table 4 Proportion of carotid lesions at different sites Sites Near wall (%) Far wall (%) In total (%) Abnormal CIMT CCA CB ICA Carotid plaques CCA CB ICA Abbreviations: CB, carotid bifurcation; CCA, common carotid artery; CIMT, carotid intima-media thickness; ICA, internal carotid artery. Table 5 CIMT indexes chosen at different sites Site Index B OR (95% CI) P CCA antcca ( ) CB meancb ( ) ICA meanica ( ) Total meanimt ( ) Abbreviations: CB, carotid bifurcation; CCA, common carotid artery; CI, confidence interval; ICA, internal carotid artery; OR, odds ratio. B stands for regression coefficient. OR equals to Exp (B). Total events All 1.87 ( ) 0.000* 1.17 ( ) Male 2.40 ( ) 0.000* 1.58 ( ) Female 1.36 ( ) ( ) MI All 1.74 ( ) 0.010* 0.77 ( ) Male 2.60 ( ) 0.004* 1.87 ( ) Female 0.88 ( ) ( ) Stroke All 1.98 ( ) 0.001* 1.43 ( ) Male 2.41 ( ) 0.004* 1.77 ( ) Female 1.61 ( ) ( ) PAD All 1.80 ( ) ( ) Male 4.31 ( ) 0.032* 3.68 ( ) Female 1.31 ( ) ( ) Abbreviations: CI, confidence interval; MI, myocardial infarction; OR, odds ratio; PAD, peripheral arterial disease. *Po0.05, when compared to subjects without events. Evaluation of CIMT at different sites Table 4 illustrates the proportion of carotid lesions (abnormal CIMT or carotid plaques) at different sites. Most lesions locate in the CB, while CCA in the second place. More lesions are found at the far walls, which might be due to the low detection rate at near walls. Logistic step-wise regression is used to assess which of the IMT indexes correlates with cardiovascular events best. In total events, after adjustment of age, the CIMT indexes chosen at each site are shown in Table 5. At CCA, mean anterior IMT correlates with total cardiovascular events best, while at other sites of measurement, mean IMT correlates with total cardiovascular events best. Taking into account all the four indexes above and maximt, meanimt is the only index finally into regression equation (regression coefficient B ¼ 1.479, OR: 4.39 ( ), P ¼ 0.000), showing that meanimt correlates with total cardiovascular events best. Similar calculations are carried out in MI and stroke. The results show that postimt is the index most strongly correlated with MI (B ¼ 1.088, OR: 2.97 ( ), P ¼ 0.000). And meanimt is most strongly associated with stroke (B ¼ 1.605, OR: 4.98 ( ), P ¼ 0.000). Discussion The association between CIMT and cardiovascular events Atherosclerosis at carotid and coronary arteries, both of which are elastic arteries, is induced by similar risk factors and pathophysiological principles. 1 Some study found that carotid and coronary atherosclerosis happened at almost the same time (strictly, carotid atherosclerosis earlier). 9 Carotid ultrasonography can be used to predict and evaluate coronary lesions. The structural and functional changes in intima are prior to plaque formation and haemodynamic changes, and represent early stages of

5 atherosclerosis. 10 CIMT can reflect atherosclerotic process directly and is an important tool in evaluating cardiovascular diseases. 4 Whether CIMT is associated with incident cardiovascular events independently of established risk factors is still in controversy. Previous studies have demonstrated inconsistent associations. The Atherosclerosis Risk in Communities study 11 and the Cardiovascular Health Study (CHS) 12 have demonstrated that CIMT is an independent predictor of cardiovascular disease after adjustment for traditional risk factors, even as powerful an indicator of cardiovascular events as the traditional risk factors. 12 However, in our study, CIMT was related to cardiovascular diseases in a univariate analysis. The subjects with cardiovascular events had significantly higher CIMT and higher abnormal CIMT proportion. The OR for total cardiovascular events comparing maximt X0.9 mm to o0.9 mm was 1.87 (P ¼ 0.000), for MI 1.74 (P ¼ 0.010) and for stoke 1.98 (P ¼ 0.001). In male subjects, the OR for total cardiovascular events comparing maximt X0.9 mm to o0.9 mm was 2.40 (P ¼ 0.000), for MI 2.60 (P ¼ 0.004) and for stoke 2.41 (P ¼ 0.004). However, after adjustment for several risk factors such as age, smoking, hypertension, diabetes and lipids, the associations were attenuated and not statistically significant, implying CIMT might act via traditional risk factors. No similar relationships were found in female subjects. Similar results were obtained in the Angina Pectoris Study in Stockholm 13 and in Rotterdam study. 14 Adding IMT to a risk function for coronary heart disease and cerebrovascular disease did not result in a substantial increase in the predictive value when used as a screening tool. 15 Now there are not sufficient evidences that CIMT might provide predictive information beyond traditional risk factors and be used in risk stratification, which might partly be due to different sites of measurements as well as actual differences in the number of risk factors included in the analyses and the age groups studied. Further prospective studies are needed. 6 There is something else to note. This cohort was of high cardiovascular risks, even in the subjects without events, 52.3% already had carotid plaques. As an index for early atherosclerotic lesions, CIMT might be more useful in subjects with moderate risks. Methods of CIMT measurement The CCA is straight, relatively superficial and parallel to the skin surface, thus good quality scans can be achieved more easily with a high precision and reproducibility rate. CCA is examined more frequently, especially with the assistance of automated computerized edge-tracking method. In contrast, the ICA is relatively difficult to be visualized. In measuring IMT of the ICA there are many missing images, and intra-observer and inter-observer variabilities are large. 16 However, CB and ICA are more prone to atherosclerosis and where plaques are most likely to form. 3 Atherosclerotic lesions appear later in the CCA than in the ICA or CB. 16 Measurement of CCA-IMT alone might lower the sensitivity of detecting atherosclerotic lesions. Both CB and coronary arteries have a turbulent blood flow and a low-oscillatory shear stress which may lead to intimal injuries and have implications in the pathogenesis of atherosclerosis in these arterial segments. 17 In our study, atherosclerotic lesions found in proximal ICA constituted a relatively small percentage, which might be due to difficulties in visualizing ICA and detecting lesions. Different site-specific imaging protocols were used in various trials. In some researches, CCA-IMT has shown to predict future cardiovascular events as well as mean CIMT. 18 Results from the Pravastatin, Lipids, and Atherosclerosis in the Carotid Arteries II even showed that IMT at the far wall of CCA was significantly correlated with progression or regression of disease other than the aggregate measurements at near and far walls of CCA, CB and ICA. 19 However, the CHS, the Rotterdam study and some other studies showed that CB-IMT was more closely related to coronary atherosclerosis, existing atherosclerotic diseases and major established risk factors compared with CCA-IMT. The combination of IMT at CCA and ICA correlates more strongly with existing cardiovascular diseases than either taken alone. 1,20,21 Bots ML et al. 18 also favoured the use of combined CIMT rather than CCA-IMT as the primary outcome measures to evaluate the efficacy of interventions in carotid atherosclerosis after a pooled analysis. In our study, the mean CIMT at multiple extracranial carotid sites (CCA, CB and ICA, near and far walls, and left and right sides) was found to be more strongly associated with cardiovascular events than the IMT from individual segments, which further ascertained the value of approaches for determining CIMT in combination. CIMT at different sites seemed to be of different value in cardiovascular diseases. CHS demonstrated that the combination of CCA-IMT and ICA (including CB)-IMT was strongly associated with the risk of MI, while the IMT of CCA was associated with stroke. 12 In this study, mean posterior CIMT correlates with MI best, while mean CIMT correlates with stroke and total events best. B-mode evaluation of the anterior walls is less reliable than the posterior walls, since according to the physical principles, the anatomic location of an interface corresponds to the image interface only when the ultrasound beam comes from a less to a more dense issue. The measurement of nearwall CIMT is also gain dependent, thus more variable. Some studies showed that anterior IMT is 80% of the histological thickness. 16 However, CIMT at anterior walls can be useful when monitored 263

6 264 in prospective studies since the variation of IMT can be reliably measured. 3 No difference was found in rates of progression between the two walls. 16 Most investigators have determined CIMT of the far walls only, while in this study CIMT of the far walls as well as the near walls were measured simultaneously. The combination of CIMT at near and far walls shows the most significance in assessing cardiovascular events. Other recent studies demonstrated that the association between near-wall CIMT and stroke or MI was as strong as that found for far-wall CIMT, and near-wall CIMT measurement can be obtained with sufficient reproducibility in a considerable proportion of the participants. 18 The combined CIMT measurements of near and far walls showed the strongest association, and may enhance precision without loss of validity. 16 In summary, although most of the carotid lesions locate at CB, the combined measurement of IMT at CCA, CB and ICA shows the most significance in prevalent cardiovascular events. CIMT at near walls alone is not suitable for assessing cardiovascular diseases, while in association with that at far walls might add informativeness. Limitations of the study The present Atherosclerosis survey is a cardiovascular epidemiological study based on a cohort from suburban community of Beijing. It evaluates the significance of CIMT in cardiovascular diseases after adjustment of traditional risk factors. It also assesses different site-specific measurements of CIMT in methodology, providing some evidences for further jobs. Nonetheless, the cross-sectional design limited our ability to draw conclusions regarding the temporality of these associations. In our study, we only compared between those people with known cardiovascular events and without. In fact, in this cohort of high cardiovascular risks, some subjects with no obvious events had already shown suspected symptoms (such as angina pectoris). It might underestimate the value of CIMT to classify these subjects into the group without events. At the same time, survival bias could not be ruled out for subjects who have undergone cardiovascular events. In the second, anti-hypertensive and lipid-lowering therapies might act as confounding factors in the evaluation of the relationship between CIMT and cardiovascular events. Finally, carotid atherosclerotic indexes other than CIMT (such as plaque size and stability, haemodynamic parameters) were not analysed. In conclusion, the study shows that an increased CIMT relates to cardiovascular events, and provides supportive evidence for the use of IMT measurements as a proxy end point in observational studies and trials. Combined IMT measurement of the near and far walls at CCA, CB and ICA might be the best index associated with cardiovascular diseases. What is known about the topic K CIMT is a good predictor of subsequent cardiovascular diseases. K Whether CIMT is associated with incident cardiovascular events independently of established risk factors is still in controversy. K There is no uniform standard as to the measurement of CIMT in clinical practice. What this study adds K Intima-media thickness measurements might be a proxy end point instead of an independent risk factor of cardiovascular diseases in observational studies and trials. K Combined IMT measurement of the near and far walls at CCA, CB and ICA might be the best index associated with cardiovascular diseases. Abbreviations: CB, carotid bifurcation; CCA, common carotid artery; CIMT, carotid intima-media thickness; ICA, internal carotid artery; IMT, intima-media thickness. References 1 O Leary DH, Polak JF, Kronmal RA, Savage PJ, Borhani NO, Kittner SJ et al. Thickening of the carotid wall: a marker for atherosclerosis in the elderly? Cardiovascular Health Study Collaborative Research Group. Stroke 1996; 27: Nissen SE. Identifying patients at risk: novel diagnostic techniques. Eur Heart J Suppl 2004; 6: C15 C20. 3 Touboul PJ. Clinical impact of intima media measurement. Eur J Ultrasound 2002; 16: Mancini GB, Dahlöf B, Díez J. Surrogate markers for cardiovascular disease: structural markers. Circulation 2004; 109: IV-22 IV Bots ML, Dijk JM, Oren A, Grobbee DE. Carotid intimamedia thickness, arterial stiffness and risk of cardiovascular disease: current evidence. J Hypertens 2002; 20: Rosvall M, Janzon L, Berglund G, Engstrom G, Hedblad B. Incident coronary events and case fatality in relation to common carotid intima-media thickness. J Intern Med 2005; 257: Eric de Groot, Hovingh GK, Wiegman A, Duriez P, Smit AJ, Fruchart JC et al. Measurement of arterial wall thickness as a surrogate marker for atherosclerosis. Circulation 2004; 109: III-33 III Simon A, Gariepy J, Chironi G, Megnien JL, Levenson J. Intima-media thickness: a new tool for diagnosis and treatment of cardiovascular risk. J Hypertens 2002; 20: Thalhammer C, Balzuweit B, Busjahn A, Walter C, Luft FC, Haller H. Endothelial cell dysfunction and arterial wall hypertrophy are associated with disturbed carbohydrate metabolism in patients at risk for cardiovascular disease. Arterioscler Thromb Vasc Biol 1999; 19: Ghiadoni L, Taddei S, Virdis A, Sudano I, Di Legge V, Meola M et al. Endothelial function and common carotid wall thickening in patients with essential hypertension. Hypertension 1998; 32: Chambless LE, Hiess G, Folsom AR, Rosamond W, Szklo M, Sharrett AR et al. Association of coronary heart disease incidence with carotid arterial wall thickness and major risk factors: the Atherosclerosis Risk in Communities (ARIC) Study, Am J Epidemiol 1997; 146:

7 12 O Leary DH, Polak JF, Kronmal RA, Manolio TA, Burke GL, Wolfson SK. Carotid-artery intima and media thickness as a risk factor for myocardial infarction and stroke in older adults. N Engl J Med 1999; 340: Held C, Hjemdahl P, Eriksson SV, Björkander I, Forslund L, Rehnqvist N. Prognostic implications of intima-media thickness and plaques in the carotid and femoral arteries in patients with stable angina pectoris. Eur Heart J 2001; 22: Bots ML, Hoes AW, Koudstaal PJ, Hofman A, Grobbee DE. Common carotid artery intima-media thickness and risk of stroke and myocardial infarction: the Rotterdam Study. Circulation 1997; 96: del Sol AI, Moons KG, Hollander M, Hofman A, Koudstaal PJ, Grobbee DE et al. Is carotid intima-media thickness useful in cardiovascular disease risk assessment? The rotterdam Study. Stroke 2001; 32: Poredos P. Intima-media thickness: indicator of cardiovascular risk and measure of the extent of atherosclerosis. Vasc Med 2004; 9: Garcia JH, Khang-Loon H. Carotid atherosclerosis: definition, pathogenesis and clinical significance. Neuroimaging Clin N Am 1996; 6: Bots ML, Evans GW, Riley WA, Grobbee DE. Carotid intima-media thickness measurements in intervention studies: design, options, progression rates, and sample size considerations: a point of view. Stroke 2003; 34: Barth JD. An update on carotid ultrasound measurement of intima-media thickness. Am J Cardiol 2002; 89: 32B 39B. 20 Lekakis JP, Papamichael CM, Cimponeriu AT, Stamatelopoulos KS, Papaioannou TG, Kanakakis J et al. Atherosclerotic changes of extracoronary arteries are associated with the extent of coronary atherosclerosis. Am J Cardiol 2000; 85: del Sol AI, Bots ML, Grobbee DE, Hofman A, Witteman JC. Carotid intima-media thickness at different sites: relation to incident myocardial infarction. Eur Heart J 2002; 23: