The relationship between mechanical properties of carotid artery and coronary artery disease

European Heart Journal Cardiovascular Imaging (2012) 13, 568 573 doi:10.1093/ejechocard/jer259 The relationship between mechanical properties of carotid artery and coronary artery disease Su-A Kim, Seong-Mi Park, Mi-Na Kim, Yong-Hyun Kim, Dong-Hyuk Cho, Chul-Min Ahn, Soon Jun Hong, Do-Sun Lim, and Wan Joo Shim* Department of Cardiology, Cardiovascular Center, Korea University Anam Hospital, 126-1, 5ga, Anamdong, Seongbukgu, Seoul 136-705, Republic of Korea Received 8 August 2011; accepted after revision 31 October 2011; online publish-ahead-of-print 28 November 2011 Aims Carotid intima-media thickness (CIMT) is a useful indicator of coronary artery disease (CAD). The relation between functional changes of the common carotid artery (CCA) and CAD is unclear. This study investigated the relationship between CCA strain and CIMT in the assessment of CAD.... Methods Within a 1-month period, 121 patients underwent coronary angiography and ultrasonography of the CCA. Circumferential strain and strain rate were measured in the left CCA by 2D ultrasonography with a speckle tracking method and results and CIMT was measured using standard techniques. Adequate images were obtained in 104 patients (59 + 11 years, 39 females), of which 46 had CAD. Strain and strain rate were negatively correlated with CIMT and were positively correlated with the number of risk factors. CIMT was significantly higher, and strain and strain rate were significantly lower in the CAD group (OR, 95% CI, P-value, CIMT: 23.23, 2.03 265.30, 0.011; strain: 0.51, 0.31 0.82, 0.006; strain rate: 0.01, 0.00 0.13, 0.001). However, after adjustment for cardiovascular risk factors, only strain and strain rate were significantly associated with CAD (OR, 95% CI, P-value, CIMT: 5.28, 0.28 99.84, 0.267; strain: 0.51, 0.30 0.88, 0.014; strain rate: 0.01, 0.00 0.42, 0.014). CAD severity, assessed by the number of stenosed coronary arteries, was also significantly correlated with strain and strain rate, but was not significantly correlated with CIMT.... Conclusion Both CIMT and CCA strain were associated with CAD. CAD severity and extent were correlated with strain and strain rate, but were not correlated with CIMT. The evaluation of mechanical properties of CCA by ultrasonographic 2D strain imaging could therefore be more effective than that by CIMT for the assessment of CAD. ----------------------------------------------------------------------------------------------------------------------------------------------------------- Keywords carotid strain carotid intima-media thickness coronary artery disease Introduction Carotid atherosclerosis is a useful indicator of cardiovascular disease. For more than 10 years, carotid artery intima-media thickness (CIMT) has been used as a clinical indicator of carotid atherosclerosis. Previous studies have reported that CIMT is associated with coronary artery disease (CAD) and cardiovascular risk factors. 1,2 CIMT reflects the focal changes in vascular morphology after the development of atherosclerosis, while vascular stiffness reflects the functional properties of carotid arteries and provides information on the changes that occur in the artery due to atherosclerosis. Conventional methods for evaluating vascular stiffness, such as the elastic modulus and distensibility coefficient, 3 are affected by the severity and progression of atherosclerosis of the coronary artery, 4 and show changes before the development of clinical symptoms or atherosclerotic plaques. 5,6 However, the validity and reproducibility of these tests are limited due to their dependence on the patient s blood pressure. 7,8 Strain imaging, using 2D ultrasonography with a speckle tracking method, records the circumferential change in the vascular morphology throughout the cardiac cycle, and directly analyses the elasticity of the carotid artery. An association between circumferential strain and the progression of vascular stiffness with age has been reported, and the superiority of strain values relative to vascular stiffness in discriminating the patient s age was observed. 9 Therefore, this study was performed to investigate the relationship between common carotid artery (CCA) strain and CIMT in the assessment of CAD. The relation between strain values CIMT, and cardiovascular risk factors and severity of CAD was * Corresponding author. Tel: +82 2 920 5448; fax: +82 2 927 1478, Email: wjshimmd@unitel.co.kr Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2011. For permissions please email: journals.permissions@oup.com

Mechanical properties of carotid artery and CAD 569 evaluated, and the predictive accuracy of strain was compared with that of CIMT for CAD. Methods Study population Among 145 subjects identified during the screening period, 24 patients were excluded on the basis of a history of previous coronary intervention,.50% carotid artery stenosis, cerebrovascular disease, decompensated heart failure, history of myocardial infarction, known vasculitis, or atrial fibrillation. In total, 121 patients who underwent elective coronary angiography with clinical suspicion of CAD, were performed carotid ultrasonography within a 1-month period. The main indication for coronary angiography was the typical chest pain suggesting CAD with positive stress test result. Informed consent was obtained from all subjects. Cardiovascular risk factors such as hypertension, diabetes mellitus, hyperlipidaemia, and smoking were evaluated by history and medical records. Hypertension was defined as blood pressure.140/90 mmhg or the use of antihypertensive medications. Diabetes mellitus was defined as fasting glucose level.126 mg/dl, random glucose level.200 mg/dl, with symptoms of hyperglycaemia, or the use of an oral hypoglycaemic agent or insulin. Hyperlipidaemia was defined as the total cholesterol level.180 mg/dl, LDL cholesterol level.100 mg/dl, HDL cholesterol level,55 mg/dl for females and,45 mg/dl for males or on statin treatment. Smoking was assessed and presented as the number of pack-years (packs smoked per day years as a smoker). Carotid ultrasonography and strain imaging Ultrasonographic images of the left carotid artery were obtained at about 1 cm proximal to the carotid bifurcation by a trained physician. The Vivid I (GE Vingmed Ultrasound) ultrasound system was used with a 14-MHz linear transducer. At least three consecutive beats were stored with a frame rate of 99 110 frames/s Strain image data were analysed with EchoPac software (GE Vingmed Ultrasound). Carotid arterial wall intima-media thickness (IMT) was evaluated by B-mode real-time ultrasound and was defined as the distance from the lumen intimal interface to the medial-adventitial border in the far artery wall. Measurement of IMT was done automatically using the automatic border detection programme in the Vivid I system. For distinction from IMT, carotid plaque was defined as a focal structure growing into the arterial lumen of at least 0.5 mm or 50% of the surrounding IMT value. The images for analysis of the patients with plaque were acquired at a location closer to the mentioned area where the plaques were not located. Circumferential strain in the carotid artery was measured by a circular region of interest (ROI) covering the cross-sectional image of the CCA. Each speckle on the carotid artery wall was identified and the movement of each speckle during the cardiac cycle was traced. Circumferential peak systolic strain (%) and early and late systolic strain rate (strain per time unit, 1/s) variables were measured as an average of the whole ROI, giving respective global strain and strain rate values as described previously 9 (Figure 1). Coronary angiography Coronary angiography was performed in all study subjects via the femoral or radial artery using standard techniques. Routinely, a minimum of two views of the right coronary artery and four views of the left coronary artery were digitally recorded. Coronary stenosis was visually estimated by comparing the luminal diameter of the segment exhibiting stenosis with the luminal diameter of the most normal-appearing site immediately proximal to the stenosis. CAD was diagnosed when any degree of luminal diameter stenosis of one of the major epicardial arteries and/or luminal wall surface irregularities was observed. The extent of stenosis was visually estimated and recorded as moderate stenosis when the luminal diameter narrowing was in the range of 70 90% and as severe stenosis when the luminal diameter stenosis was.90% due to an occlusive lesion with preserved coronary blood flow judged visually. The severity of CAD was determined by assessing the number of major epicardial arteries with stenosis. Statistical analysis Statistical analysis was performed with SPSS, version 12. Data are expressed as mean + SD and a P-value of 0.05 was considered significant. Differences in clinical data, strain variables, and common CIMT between patients with and without CAD were tested with the x 2 - test and unpaired t-tests. The linear correlations among strain variables, CIMT, and patient age were evaluated by Pearson s correlation analysis. Analysis of variance (ANOVA) of CIMT, strain and strain rate was performed in conjunction with the number of cardiovascular risk factors. A stepwise logistic regression analysis was performed to assess the influence of the cardiovascular risk factors including sex, age, hypertension, diabetes mellitus, hyperlipidaemia and smoking on strain variables and CIMT. The association of severity of CAD with CIMT and strain variables was assessed with ANOVA. Results Among 121 patients, 17 patients were excluded due to inadequate images for analysis. The remaining 104 patients were divided into the CAD group and the non-cad group according to coronary angiography findings. The clinical characteristics of both groups are shown in Table 1. There were no significant differences in the male to female ratio between the two groups, but the patients with CAD were older than the patients without CAD. The mean age of the non-cad group was 57 + 12 years and that of the CAD group was 62 + 9 years (P ¼ 0.042). There were no significant differences in the prevalence of hypertension or diabetes mellitus between groups. There were more cases of hyperlipidaemia and more smokers in the CAD group than in the non-cad group (Table 1). The results of the analysis irrespective of CAD showed a direct correlation between CIMT and strain values. Patients who had higher CIMT had lower strain and strain rate values (strain: r ¼ 20.196, P ¼ 0.046; strain rate: r ¼ 20.224, P ¼ 0.022). Patient age was significantly correlated with both CIMT and strain values. Older patients demonstrated higher CIMT (r ¼ 0.542, P, 0.001) and lower strain values (strain: r ¼ 20.352, P, 0.001; strain rate: r ¼ 20.382, P, 0.001; Figure 2). Mean values of CIMT did not differ significantly with the number of risk factors, whereas those of strain and strain rate were significantly different (CIMT: P ¼ 0.124, strain: P ¼ 0.017, strain rate: P ¼ 0.036). Table 2 and Figure 3 show the mean values of CIMT, strain, and strain rate of the carotid artery in the CAD and the non-cad groups. CIMT was significantly higher, and strain values were

570 S.A. Kim et al. Figure 1 One example of the global strain (upper) and strain rate of carotid artery (lower). The peak value of strain and strain rate were adopted in this study. Table 1 Baseline characteristics of the study population Non-CAD CAD P-value... Number 58 46 Age (years) 57 + 12 62 + 9 0.042 Sex (M/F) 34/24 31/15 0.418 Diabetes mellitus (%) 24.6 20.5 0.602 Hypertension (%) 46.4 58.7 0.238 Smoking (%) 33.3 39.1 0.680 Dyslipidaemia (%) 69.0 91.3 0.007 SBP (mmhg) 130 + 16 129 + 16 0.820 DBP (mmhg) 81 + 12 77 + 10 0.123 HR (bpm) 73 + 14 75 + 16 0.524 BMI (kg/m 2 ) 24.9 + 2.8 24.5 + 2.3 0.602 Medication ACEi or ARB 19.0 26.1 0.477 BB 15.5 21.7 0.452 CCB 25.9 13.0 0.141 SBP, systolic blood pressure; DBP, diastolic blood pressure; HR, heart rate; BMI, body mass index; ACEi, angiotensin converting enzyme inhibitor; ARB, angiotensin receptor blocker; BB, beta blocker; CCB, calcium channel blocker. significantly lower in the CAD group. Mean CIMT was 0.72 + 0.17 mm in the non-cad group and 0.82 + 0.16 mm in the CAD group (P ¼ 0.009). The mean values of strain and strain rate were 2.81 + 0.91% and 0.45 + 0.17 in the non-cad group and 2.29 + 0.84% and 0.34 + 0.11 in the CAD group (P ¼ 0.004, P, 0.001), respectively. Odds ratios for CIMT and strain values for prediction of CAD were determined by logistic regression analysis. The odds ratios of CIMT, strain, and strain rate were 23.229, 0.511, and 0.005, respectively, with significant confidence intervals (2.034 256.308, 0.317 0.822, 0.000 0.130, respectively) and P-values (0.011, 0.006, 0.001, respectively). However, when risk factors including age, sex, diabetes mellitus, hypertension, smoking, dyslipidaemia were adjusted, the odds ratio for CIMT was not valid, whereas odds ratios of strain values were significant (odds ratio, 95% CI, P-value: CIMT: 5.280, 0.279 99.842, 0.267; strain: 0.511, 0.298 0.875, 0.014; strain rate: 0.013, 0.000 0.415, 0.014, Table 3). Disease severity was more strongly correlated with strain and strain rate than with CIMT. When the severity of CAD was graded using the number of major epicardial arteries with severe stenosis, there were significant differences in the mean CIMT and strain values in each severity group. However, when the severity of CAD was graded using the number of vessels with more than

Mechanical properties of carotid artery and CAD 571 Table 2 Carotid intima-media thickness and strain values in patients with and without coronary artery disease Non-CAD CAD P-value... CIMT (mm) 0.72 + 0.17 0.82 + 0.16 0.009 Strain (%) 2.81 + 0.91 2.29 + 0.84 0.004 Strain rate (1/s) 0.45 + 0.17 0.34 + 0.11,0.001 Figure 2 Correlation analysis of age with carotid intima-media thickness (A), strain (B), and strain rate (C) with age. moderate stenosis, only strain and strain rate were significantly different between severity groups (Table 4). Discussion This study evaluated the clinical characteristics of ultrasonographic speckle tracking-based strain values and their relationship with CIMT in the context of CAD in patients who underwent coronary angiography and carotid artery ultrasonography within 1 month among the patients with symptoms suggesting CAD and positive non-invasive test result for CAD. The relation between carotid strain and cardiovascular risk factors was also examined. We found that lower strain values were associated with advancing age, greater number of cardiovascular risk factors, higher CIMT, and CAD. The correlation of strain values with both the number of cardiovascular risk factors and the odds ratio for predicting CAD after adjustment for CAD risk factors was stronger than that of CIMT values, which did not reach statistical significance. The association of strain values with the severity of CAD was also stronger than that for CIMT values. Many studies have tried to present arterial stiffness as a manifestation of atherosclerotic cardiovascular disease. 5,6 All-cause and cardiovascular mortality have been noted to be correlated with arterial stiffness 6 and the assessment of arterial stiffness is increasingly used in clinical practice. Brachial-ankle pulse-wave velocity (PWV) and distensibility coefficient, or stiffness parameter b of the carotid artery or aorta, are the methods widely used for obtaining objective information on arterial stiffness. However, there are limitations to these techniques. Brachial-ankle PWV is influenced by the patient s blood pressure. Stiffness parameter b, which is determined by the change in the vascular diameter on M-mode ultrasonography, 10 is a one-dimensional approach having various limitations, and the influence of patient s blood pressure also cannot be avoided. 11 Of late, strain values have enabled the researchers to assess the elastic properties of arterial wall directly. 12,13 Ultrasonographic 2D imaging or tissue Doppler imaging was performed throughout the cardiac cycle, and the change in the circumference or thickness of arterial wall was analysed. This simple and direct process showed good reproducibility and the limitations of the previous methods were avoided. Arterial wall elasticity, assessed with strain imaging, was also identified as a possible marker of atherosclerosis. 9 There are few reports describing the clinical implications of strain values. Bjallmark et al. 9 reported that age-dependent differences in the elastic properties of the CCA could be successfully

572 S.A. Kim et al. Figure 3 Median and average of carotid intima-media thickness (A) and strain (B) and strain rate (C) in patients with and without coronary artery disease. Table 3 Logistic regression analysis of carotid intima-media thickness and strain values in the presence of coronary artery disease Unadjusted... Risk factors a adjusted... Confidence interval Odds ratio P-value Confidence interval Odds ratio P-value... CIMT (mm) 2.034 265.308 23.229 0.011 0.279 99.842 5.280 0.267 Strain (%) 0.317 0.822 0.511 0.006 0.298 0.875 0.511 0.014 Strain rate (1/s) 0.000 0.130 0.005 0.001 0.000 0.415 0.013 0.014 a Risk factors include age, sex, hypertension, diabetes mellitus, hyperlipidaemia, and smoking. Table 4 Values of carotid intima-media thickness and strain according to severity of coronary artery stenosis No stenosis 1 vessel 2 vessel 3 vessel P-value... Stenosis.70% CIMT (mm) 0.74 + 0.17 0.82 + 0.15 0.83 + 0.17 0.80 + 0.23 0.192 Strain (%) 2.76 + 0.89 2.17 + 0.90 2.41 + 0.85 1.77 + 0.57 0.009 Strain rate (1/s) 0.43 + 0.16 0.35 + 0.14 0.32 + 0.07 0.28 + 0.08 0.018 Stenosis.90% CIMT (mm) 0.75 + 0.16 0.81 + 0.16 0.89 + 0.18 0.54 0.03 Strain (%) 2.71 + 0.90 2.30 + 0.86 1.85 + 0.74 2.34 0.032 Strain rate (1/s) 0.42 + 0.16 0.36 + 0.13 0.28 + 0.06 0.37 0.048 evaluated with strain values acquired using a speckle tracking method. Their data showed the superiority of strain values compared with conventional stiffness variables. Oishi et al. 12 reported a correlation of age with the circumferential strain of the aorta. Strain values showed a closer correlation than b stiffness. Finally, Kawasaki et al. 13 found that carotid artery radial strain values

Mechanical properties of carotid artery and CAD 573 obtained by tissue Doppler imaging correlated with patient age, Framingham risk score and the results of coronary angiography. However, to the best of our knowledge, to date no report has demonstrated the possibility of predicting CAD with strain values assessed by 2D strain imaging. In this study, the statistically insignificant result of CIMT for predicting CAD after adjustment of cardiovascular risk factors was different from that in most of the current reports. 4,14 This may be due to an insufficient number of patients diagnosed with CAD, and the fact that the group of patients without CAD did not consist of normal healthy patients. There were no group differences in the prevalence of hypertension, diabetes mellitus and smokers, which could have influenced the CIMT values. 2 However, CAD was successfully assessed with strain and strain rate after adjustment of cardiovascular risk factors under the same conditions, which shows a strong possibility that strain values could represent a marker for CAD. The CAD group showed a higher prevalence of dyslipidaemia, smokers and older patients. Such risk factors were adjusted using a logistic regression analysis, and the result was statistically significant. The risk factors did not influence the results of this study. CIMT and strain values were measured in the left CCA. Some reports suggest differences between the left and right CIMTs. 15 17 This difference is explained to be as a result of the different effects of hemodynamic stress on the left and right CCA, due to their different anatomical origins. The left CCA originates from the aortic arch, and is affected by the hydrostatic pressure of the aortic arch directly. However, the right CCA originates from the brachiocephalic (innominate) artery, and is affected by dynamic pressure from ascending aortic blood flow. 15 Luo et al. 15 reported that left CIMT was thicker and more sensitively discriminated the age of patients. Hernandez et al. 16 reported the superiority of left-side measurements in predicting cerebrovascular disease. There are some limitations to this study. As stated above, the subjects of this study were patients who underwent CAG due to clinical suspicion of CAD, and there were no normal controls. The characteristics of strain values in this study, or mean values of strain and strain rate are difficult to represent in the general population. Moreover, only 39 patients were diagnosed with CAD. The statistical significance for the correlation between the severity of CAD and strain values or CIMT could be weak, especially in the CAD group. We also tried to evaluate the influence of left ventricular systolic function. There were no group differences in strain values between the patients with normal systolic function and diminished systolic function (,50% EF). The odds ratio and 95% CI of strain values were still significant when adjusted for left ventricular systolic function. However, the number of patients with diminished systolic function was 14 out of 104 patients, and hence it was difficult to evaluate the influence of left ventricular systolic function on strain and strain rate. Finally, the fact that 17 of 121 included patients failed to complete an adequate analysis for strain and strain rate should be taken into consideration. For an accurate analysis, precise images were needed, and patients with poor image quality were excluded. Especially, the individual anatomical variances such as short-neck length or CCA adjacent to other vessels were the main causes of poor image quality. In conclusion, the findings of this study suggest that lower carotid strain values are associated with the presence and the severity of CAD. Strain imaging using a speckle tracking method successfully assessed carotid artery elasticity directly, and was superior to CIMT in predicting CAD. Therefore, measuring the extent of carotid artery wall motion and deformation behaviour by strain imaging has the potential to provide useful information on CAD. Conflict of interest: none declared. References 1. 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;340:14 22. 2. Coskun U, Yildiz A, Esen OB, Baskurt M, Cakar MA, Kilickesmez KO et al. Relationship between carotid intima-media thickness and coronary angiographic findings: a prospective study. Cardiovasc Ultrasound 2009;7:59. 3. Kawasaki T, Sasayama S, Yagi S, Asakawa T, Hirai T. Non-invasive assessment of the age related changes in stiffness of major branches of the human arteries. Cardiovasc Res 1987;21:678 87. 4. Lee MY, Wu CM, Yu KH, Chu CS, Lee KT, Sheu SH et al. Association between hemodynamics in the common carotid artery and severity of carotid atherosclerosis in patients with essential hypertension. Am J Hypertens 2008;21:765 70. 5. Laurent S, Boutouyrie P. Arterial stiffness: a new surrogate end point for cardiovascular disease? J Nephrol 2007;20:45. 6. Mattace-Raso FUS, van der Cammen TJM, Hofman A, van Popele NM, Bos ML, Schalekamp MADH et al. Arterial stiffness and risk of coronary heart disease and stroke: the rotterdam study. Circulation 2006;113:657. 7. Pannier BM, Avolio AP, Hoeks A, Mancia G, Takazawa K. Methods and devices for measuring arterial compliance in humans. Am J Hypertens 2002;15:743 53. 8. Hamilton PK, Lockhart CJ, Quinn CE, McVeigh GE. Arterial stiffness: clinical relevance, measurement and treatment. Clin Sci (Lond) 2007;113:157 70. 9. Bjallmark A, Lind B, Peolsson M, Shahgaldi K, Brodin LA, Nowak J. Ultrasonographic strain imaging is superior to conventional non-invasive measures of vascular stiffness in the detection of age-dependent differences in the mechanical properties of the common carotid artery. Eur J Echocardiogr 2010;11:630 6. 10. Mizuguchi Y, Tanaka H, Oishi Y, Miyoshi H, Emi S, Ishimoto T et al. Predictive value of associations between carotid arterial sclerosis and left ventricular diastolic dysfunction in patients with cardiovascular risk factors. J Am Soc Echocardiogr 2007;20:806 12. 11. Lehmann E, Hopkins K, Jones R, Rudd A, Gosling R. Aortic distensibility in patients with cerebrovascular disease. Clin Sci(Lond) 1995;89:247 53. 12. Oishi Y, Mizuguchi Y, Miyoshi H, Iuchi A, Nagase N, Oki T. A novel approach to assess aortic stiffness related to changes in aging using a two-dimensional strain imaging. Echocardiography 2008;25:941 5. 13. Kawasaki T, Fukuda S, Shimada K, Maeda K, Yoshida K, Sunada H et al. Direct measurement of wall stiffness for carotid arteries by ultrasound strain imaging. J Am Soc Echocardiogr 2009;22:1389 95. 14. Di Bello V, Carerj S, Perticone F, Benedetto F, Palombo C, Talini E et al. Carotid intima-media thickness in asymptomatic patients with arterial hypertension without clinical cardiovascular disease: relation with left ventricular geometry and mass and coexisting risk factors. Angiology 2010;60:705 13. 15. Luo X, Yang Y, Cao T, Li Z. Differences in left and right carotid intima media thickness and the associated risk factors. Clin Radiol 2011;66:393 398. 16. Hernandez SAR. Is there a side predilection for cerebrovascular disease? Hypertension 2003;42:56 60. 17. Touboul PJ, Hennerici MG, Meairs S, Adams H, Amarenco P, Bornstein N et al. Mannheim carotid intima-media thickness consensus (2004 2006). Cerebrovasc Dis 2007;23:75 80.