- Email: [email protected]
The utility of ultrasonic tissue characterization of carotid plaque in the prediction of cardiovascular events in diabetic patients
Atherosclerosis 230 (2013) 399e405
Contents lists available at ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
The utility of ultrasonic tissue characterization of carotid plaque in the prediction of cardiovascular events in diabetic patients Yoko Irie a, Naoto Katakami a, b, *, Hideaki Kaneto a, Mitsuyoshi Takahara a, Mayu Nishio c, Ryuichi Kasami a, Ken’ya Sakamoto c, Yutaka Umayahara d, Satoru Sumitsuji e, Yasunori Ueda c, Keisuke Kosugi c, Iichiro Shimomura a a
Department of Metabolic Medicine, Osaka University Graduate School of Medicine, Japan Department of Metabolism and Atherosclerosis, Osaka University Graduate School of Medicine, Japan Osaka Police Hospital, Japan d Osaka General Medical Center, Japan e Department of Advanced Cardiovascular Therapeutics/Cardiology, Osaka University Graduate School of Medicine, Japan b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 April 2013 Received in revised form 15 August 2013 Accepted 15 August 2013 Available online 26 August 2013
Objective: The aim of this study was to evaluate whether non-invasive ultrasonic tissue characterization of carotid plaque using gray-scale median (GSM) can be a predictor of future cardiovascular disease (CVD) events in type 2 diabetic patients. Methods: A total of 287 type 2 diabetic patients with carotid plaque but without CVD were enrolled (male 72%, mean age 65 7 years). We prospectively evaluated the association between GSM, a quantitative parameter of the plaque echogenicity, and CVD. Results: The median follow-up period was 55 months, and there were 34 new CVD events. The risk of CVD event was significantly higher in the patients with echolucent (GSM 37) plaque (n ¼ 67) as compared to those without (n ¼ 220) (HR ¼ 6.99, 95% CI 3.46e14.14, p < 0.001). Cox proportional hazards regression analysis showed that the presence of echolucent plaque (HR ¼ 4.55, 95% CI 2.10e19.84, p < 0.001) as well as plaque thickness (HR ¼ 1.44, 95% CI 1.01e2.06, p ¼ 0.005) were independent predictors of CVD, even after adjustment for other risk factors. Time-dependent receiver-operatingcharacteristic curve analysis revealed that the addition of plaque thickness to Framingham risk score (FRS) resulted in significant increase in area under the curve (AUC) [from 0.60 (95% CI; 0.49e0.70) to 0.73 (95% CI; 0.63e0.82), p < 0.05]. Notably, the addition of plaque echogenicity (presence/absence of echolucent plaque) to the FRS and plaque thickness resulted in further and significant increase in AUC [from 0.73 (95% CI; 0.63e0.82) to 0.82 (95% CI; 0.75e0.88), p < 0.05]. Conclusion: Ultrasonic tissue characterization of carotid plaque using the GSM can improve the risk prediction of cardiovascular event in asymptomatic type 2 diabetic patients with carotid plaque. Ó 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Carotid arteries Cardiovascular diseases Diabetes mellitus Vulnerable plaque Carotid ultrasound
1. Introduction Since cardiovascular disease (CVD) remains the main cause of death and impairment of quality of life in diabetic patients, early identification of individuals at high risk for CVD events and subsequent rapid interventions are important. Recently, ultrasound examination of the carotid artery has been in use to non-invasively
* Corresponding author. Department of Metabolic Medicine, Osaka University Graduate School of Medicine, 2-2, Yamadaoka, Suita, Osaka 565-0871, Japan. Tel.: þ81 6 6879 3743; fax: þ81 6 6879 3739. E-mail address: [email protected] (N. Katakami). 0021-9150/$ e see front matter Ó 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atherosclerosis.2013.08.015
identify individuals at high risk for CVD. Some studies have shown that the presence/absence of carotid plaque assessed by ultrasound could improve the prediction ability for CVD over and above conventional coronary risk factors [1e5]. It is well known that disruption of an atherosclerotic plaque plays a crucial role in the pathogenesis of CVD event and that plaque disruption is dependent on the content of lipid, neovascular vessel, and inflammatory cells in the atheroma, and the thickness of the fibrous cap [6e9]. Therefore, tissue characterization of a plaque lesion is considered to be useful for identifying patients at high risk for CVD and death. Presently, various modalities such as Computed Tomography (CT), Magnetic Resonance Imaging (MRI), or Fluorodeoxyglucose-Positron Emission Tomography (FDG-PET) are
400
Y. Irie et al. / Atherosclerosis 230 (2013) 399e405
being used to detect vulnerable plaque in coronary and carotid arteries [10e12]. However, it is unrealistic to screen all the asymptomatic patients for individuals at high risk for CVD with these tools, since these tests are limited by the potential of significant adverse effects, technical difficulty, availability, and cost. Carotid arteries as well as the aorta develop foam cell lesions and lipid core plaque at an early age [13]. Interestingly, a recent study reported that the prevalence of lipid-rich plaques in carotid arteries was higher in coronary death than in non-coronary death [14]. Another study using carotid endarterectomy specimens also reported that a soft carotid plaque with a large necrotic core and a marked inflammatory component may be predictive of an increased risk of cerebrovascular events [15]. These findings indicate that the individuals who have vulnerable plaques in their carotid arteries are prone to also have vulnerable plaques in their coronary and/or cerebral arteries, which will lead to CVD events, and thus, the assessment of tissue characteristics of carotid plaques will help to identify the patients with a high risk for CVD event. Indeed, based on a prospective study in 85 asymptomatic type 2 diabetic patients, we have already shown that a quantitative ultrasonic tissue characterization of carotid plaque using integrated backscatter (IBS) analysis can be a predictor of future CVD event [16]. However, the measurement of IBS values requires a specific software package, “Acoustic Densitometry,” installed in a SONOS echocardiograph system (Philips Medical Systems), which would limit the widespread use of this approach. In a qualitative evaluation using conventional B-mode ultrasound imaging performed by experienced sonographers, vulnerable plaques, which consist mainly of neovascular vessel, high lipid content, and inflammatory infiltration [8,9], are believed to appear “hypoechoic.” On the other hand, stable plaques, which consist mainly of fibrous tissue and calcific components, appear “hyperechoic” [17]. Recently, echogenicity of carotid plaques has become semi-quantitatively evaluable with high-resolution B-mode ultrasound and computer-assisted image processing: the gray-scale median (GSM) of the frequency distribution of gray values of the pixels within the plaque serves as a measure of echogenicity. Since the evaluation of GSM values in carotid plaques requires neither specific software nor a specific ultrasonograph, it can be a more universal and practical approach as compared with IBS analysis. Based on the background described above, the current study prospectively evaluated whether non-invasive and inexpensive ultrasonic tissue characterization of carotid plaque using GSM can improve the risk prediction of CVD event in asymptomatic type 2 diabetic patients with carotid plaque. 2. Research design and methods 2.1. Subjects Middle-aged and older Japanese type 2 diabetic patients with carotid plaque but without apparent CVD participated in this study. Patients who fulfilled the following criteria were considered eligible: (1) age 40 years at the time of enrollment, and (2) diagnosed with type 2 diabetes based on the criteria of the Japan Diabetes Society. Exclusion criteria were (1) history of ischemic stroke, coronary heart disease, or peripheral artery disease; (2) elevated liver enzymes (GOT or GPT 2.5 times the normal range); and (3) renal insufficiency (serum creatinine 2.0 mg/dL). Screening of the study patients was performed consecutively during the registration period (from April 2007 to December 2009) at the outpatient diabetes clinic of Osaka Police Hospital. All the patients that met eligibility criteria were asked whether they could participate in the present study, and all the patients who agreed to participate were registered. Out of 325 patients who had been
registered, four were excluded because of poor image or acoustic shadowing due to severe calcification, and 34 were excluded because of no plaque. After the exclusions, a total of 287 patients were analyzed in this study. This study was conducted in agreement with the principles of the Helsinki declaration. The study protocol was approved by the committee on human research ethics of Osaka Police Hospital. Written informed consent was obtained from all the participants after a full explanation of the study. 2.2. Clinical and biochemical analyses Blood samples were collected after an overnight fast for analysis of serum concentrations of glucose, total cholesterol, triglycerides, HDL cholesterol, LDL cholesterol, and HbA1c. Diagnosis of type 2 diabetes was made by the principal investigator in accordance with the standards of the Japan Diabetes Society [18]. Dyslipidemia was defined as LDL cholesterol 120 mg/dL, HDL cholesterol < 40 mg/ dL, triglycerides 150 mg/dL, or current use of lipid-lowering agents in accordance with the standards of the Japan Diabetes Society [18]. Hypertension was defined as systolic blood pressure 130 mmHg and/or diastolic blood pressure 80 mmHg or current use of antihypertensive agents. The risk of CVD was estimated using the Framingham D’Agostino equations [19]. Risk factors included in the model to assess risk of CVD were gender, age, total cholesterol, HDL cholesterol, systolic blood pressure, antihypertensive medication use, current smoking, and diabetes status. 2.3. Ultrasound examination B-mode ultrasonography of the carotid artery was performed using an ultrasound machine (Toshiba SSA-790CE; Toshiba Medical Systems, Tokyo, Japan) with a 7.5-MHz linear transducer. All scanning was conducted by experienced laboratory physicians using the same ultrasound system and the same measuring method. Scanning of the extracranial common carotid artery (CCA), the carotid bulb (CB), and the internal carotid artery (ICA) in the neck was performed bilaterally in three different longitudinal projections and three different transverse projections. According to the guideline of the Japan Society of Ultrasonics [20], carotid plaque was defined as a focal structure encroaching into the arterial lumen or demonstrating a thickness > 1.0 mm as measured from the mediaeadventitia interface to the intra-lumen interface. The plaque thickness in the CCA, the CB, and the ICA were measured separately, and the greatest value among them was used as the representative value for each individual. Reproducibility analysis of 20 replicate measurements yielded absolute mean differences of 0.03 0.05 mm for plaque thickness. The inter-observer coefficient of variation for measurement of plaque thickness was 0.7%. The plaque echogenicity was evaluated based on the gray-scale median (GSM) in a gray-scale range of 0e255 (0 as the darkest and 255 as the brightest tone). Adobe Photoshop software (Adobe Systems, version 7.0, San Jose, CA, USA) was used for image standardization and calculation of gray-scale values. According to the criteria reported previously, the standardization of the B-mode image was performed by using a curve option, so that the GSM for the blood ranged from 0 to 5 and for the adventitia from 185 to 195 [21]. Thus the gray-scale values of all pixels would change according to the new linear scale defined by the reference values for blood and adventitia. The plaque was then delineated with a freehand tool, and the GSM of each plaque was read from the entire delineated area. In case there were multiple plaques in one individual, all the plaques present were subject to GSM measurement and the lowest value among them was used as the representative
Y. Irie et al. / Atherosclerosis 230 (2013) 399e405
value for the patients. To avoid inter-reader variability, all scans were electronically stored, sent to a central office, and read in a random order by a single reader (Y. I.) unaware of the clinical characteristics of the patients. The intra-reader coefficient of variation for GSM measurements was 5.8% for 140 consecutively replicated measurements (male 70%, age 67 7.3 years). 2.4. Assessment of cardiovascular events The prespecified primary endpoint was the first occurrence of fatal or non-fatal CVD during the follow-up period, which was a composite of any coronary heart disease (angina [both unstable and stable], myocardial infarction [both ST elevation myocardial infarction (STEMI) and non-ST elevation myocardial infarction (NSTEMI)], and coronary revascularization), any ischemic stroke (transient ischemic attack, fatal and non-fatal ischemic stroke), and peripheral artery disease during the follow-up period. The diagnosis of the occurrence of coronary heart disease was performed by cardiologists based on the clinical symptoms, characteristic electrocardiogram changes, cardiac enzyme levels, and the findings from coronary angiography and/or echocardiography, according to established guidelines. Indication for coronary revascularization for patients with stable coronary artery disease was decided based on ACC/AHA guidelines for percutaneous coronary intervention [22]. An ischemic stroke event was defined as a validated definite or probable hospitalized atherothrombotic, cardioembolic, lacunar, or other type of ischemic stroke diagnosed by neurosurgical experts based on clinical symptoms and neuroimaging findings, according to the National Institute of Neurologic Disorders and Stroke III classification. Peripheral artery disease was defined as patients with intermittent claudication due to atherosclerosis of the lower extremity arteries; evaluated based on ankle-brachial index. For participants with incident CVD event, follow-up was defined as the period between the baseline clinic visit and the date of the first event. For participants with no CVD event, follow-up continued until the date of death or December 11th 2012 or until the date of last contact. All the study patients were recommended to visit Osaka Police Hospital for treatment every 1e2 months throughout the follow-up period. At each hospital visit, medical interview, physical examination, and blood examination were performed, and the occurrence of CVD events was then determined. For a potential new CVD event, even if it was recognized at another medical institution, additional information, including the results of imaging and other diagnostic procedures, was obtained for confirmation, and all causes of death were confirmed by medical certificate. 2.5. Statistical analyses All statistical analyses were performed with SPSS version 11.0J for Windows (SPSS, Inc., Japan). All values are reported as mean SD, median (IQR), or real number of patients with the percentage in parentheses. Cox proportional hazards regression model was used to determine the unadjusted association of each variable with the outcome and hazard ratios (HRs) and 95% CIs were reported. The statistically significant variables in the unadjusted analyses were entered into multivariate models to reveal the independent impact on the outcome. The prediction ability of variables to predict CVD was examined by time-dependent receiver-operating-characteristic (ROC) curve analysis [23]. Curves were generated based on models of the prediction of risk with the use of Framingham risk score (FRS) alone, or with plaque thickness, or with plaque thickness and plaque echogenicity (presence/ absence of echolucent plaque). For all tests, p < 0.05 was considered statistically significant.
401
3. Results 3.1. Patient characteristics The baseline characteristics of the study population are shown in Table 1. Among a total of 287 patients (males, 72%; age, 65 7 years (mean SD); diabetes duration, 13 9 years; HbA1c, 7.5 1.1%.), 236 (82%) patients had hypertension and 215 (75%) had dyslipidemia. During a median follow-up of 55 months, 34 of the 287 patients (12%) had new CVD events: 11 had ischemic stroke, 21 had coronary heart disease, and 2 had peripheral artery disease (Supplementary Table 1). During the observation period, 4 participants died (2 were cardiac sudden deaths and the other 2 were cancer deaths). As compared to the patients who did not develop CVD during the observation period, the patients who developed CVD were significantly more likely to be male, and showed significantly higher smoking rates and HbA1c values (Table 1). There was no significant difference in the baseline therapeutic regimens for diabetes, dyslipidemia, or hypertension between the patients who developed CVD and those who did not. The baseline prescription rate of anti-aggregating agents was significantly greater in patients with CVD events. 3.2. Association between plaque echogenicity and cardiovascular events in patients with type 2 diabetes mellitus At baseline, plaque thickness was significantly greater (2.98 1.05 mm with CVD events vs. 2.16 0.79 mm without, p < 0.001) and plaque number was significantly greater (4.4 2.0 with CVD events vs. 3.4 1.8 without, p ¼ 0.005) in patients who developed CVD events during the observation period as compared to those who did not (Table 1). The risks for CVD events were also compared among the quartiles of plaque thickness and plaque number, and it was confirmed that both plaque thickness and plaque number showed a positive relationship with the risk of CVD events. Compared to the lowest quartile of plaque thickness (1.5 mm), the highest quartile (2.8 mm) had a significantly increased risk. Compared to the lowest quartile of plaque number (plaque number ¼ 1 or 2), the 3rd (4) and the highest quartile (5) had a significantly increased risk (Fig. 1a and b). GSM value was significantly lower in patients with CVD events as compared to those without (38 16 with CVD events vs. 51 17 without, p < 0.001). In 184 (64%) out of 287 cases, the thickest plaque also showed the lowest GSM value. First, to evaluate the relationship between the echogenicity in carotid plaque and the occurrence of CVD events in detail, the GSM dataset was divided into quartiles (Q1: 59, Q2: 48e58, Q3: 38e47, and Q4: 37) and the decrement of GSM level and the risk for CVD events were analyzed. The HRs and 95% CIs of Q2, Q3, and Q4 relative to Q1 were 0.54 (95% CI 0.13e2.24; p ¼ 0.393), 0.75 (95% CI 0.20e2.81; p ¼ 0.674), and 5.28 (95% CI 2.00e13.94; p ¼ 0.001), respectively (Fig. 1c). Thus, the lowest GSM quartile (Q4: GSM 37) had significantly and much higher risk for CVD as compared to the other GSM quartiles, and the association between GSM values and the risk of CVD was not linear. Therefore, the subsequent statistical analyses were performed based on the hypothesis that carotid plaques with GSM value 37 are considered to be clinically “echolucent” plaques. When the study patients were divided into two groups (patients with low (37) GSM plaques (n ¼ 67) and those with high (>37) GSM plaques (n ¼ 220)), the risk of cardiovascular event was significantly higher in the former as compared to the latter (HR 6.99, 95% CI 3.46e14.14; p < 0.001). Similarly, patients with echolucent (GSM 37) plaque had significantly higher risk for ischemic stroke (HR 5.86, 95% CI 1.72e20.00;
402
Y. Irie et al. / Atherosclerosis 230 (2013) 399e405
Table 1 Baseline characteristics of the study population. All subjects (n ¼ 287)
Cardiovascular events
p Value
(þ) (n ¼ 34)
() (n ¼ 253)
Age (years) Male Smoking habits (yes/no) Diabetes duration (years) BMI (kg/m2) HbA1c (%) Creatinine (mg/dL) Hypertension Dyslipidemia Plaque number Plaque thickness (mm) GSM
65 7 207 (72) 161 (56) 10 (7e17) 24 3 7.5 1.1 0.9 0.2 236 (82) 215 (75) 3.5 1.9 2.26 0.86 49 17
67 8 30 (88) 27 (79) 13 (7e21) 24 3 7.9 1.3 0.9 0.3 28 (82) 27 (79) 4.4 2.0 2.98 1.05 38 16
65 7 177 (70) 134 (53) 10 (7e17) 24 3 7.5 1.1 0.8 0.2 208 (82) 188 (74) 3.4 1.8 2.16 0.79 51 17
NS 0.025 0.005 NS NS 0.016 NS NS NS 0.005 <0.001 <0.001
Treatment approach for diabetes Diet only Insulin treatment Oral anti-hyperglycemic agents
13 (5) 89 (31) 232 (81)
3 (9) 13 (38) 26 (76)
10 (4) 76 (30) 206 (81)
NS NS NS
Achievement of targeted values of blood pressure
129 (45)
16 (47)
113 (45)
NS
Treatment approach for hypertension Using antihypertensive drugs Calcium-channel blockers ACEI or ARB
187 (65) 81 (28) 150 (52)
22 (65) 8 (24) 19 (56)
165 (65) 73 (29) 131 (52)
NS NS NS
Achievement of targeted values of LDL-C levels
179 (62)
20 (59)
159 (63)
NS
Treatment approach for dyslipidemia Statins
108 (38)
12 (35)
96 (38)
NS
Using anti-aggregating agents
60 (21)
18 (53)
42 (17)
<0.001
Abbreviation: BMI, body mass index; GSM, gray-scale median; LDL-C, low-density lipoprotein cholesterol; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker. Based on the guideline of the Japan Diabetes Society, the targeted values of blood pressure and LDL-C levels were set at <130/80 mmHg and <120 mg/dL, respectively. All values are reported as mean SD, median (IQR), or real number of subjects with the percentage in parentheses. p Values compared with diabetes patients with and without cardiovascular events by unpaired-t, ManneWhitney U, or Chi square test.
p ¼ 0.005) and coronary heart disease (HR 7.13, 95% CI 2.88e17.68; p < 0.001), respectively. Next, a multivariate Cox proportional hazard regression model was used, in which HbA1c, Framingham risk score (FRS), plaque thickness, plaque number, the use of anti-aggregating drugs, and the presence of echolucent plaque were included as independent variables. This analysis revealed that the presence of echolucent plaque (HR 4.55 [95% CI 2.10e19.84]; p < 0.001) as well as plaque thickness (HR 1.44 [1.01e2.06]; p ¼ 0.005) were independently associated with CVD, even after adjustment for other factors (Table 2). To confirm that the presence of echolucent plaque was an independent predictor of CVD events even after adjustment for renal function and the treatment regimen, other multivariate Cox proportional hazard regression models were also tested. These analyses showed that the presence of echolucent plaque was an independent predictor of CVD events, even after adjustment for serum creatinine levels and the treatment regimen at baseline (data not shown). 3.3. Contribution of GSM in the prediction of cardiovascular event in patients with type 2 diabetes mellitus To examine whether the addition of plaque echogenicity and/or plaque thickness to conventional risk factors could improve the prediction ability for CVD event, time-dependent ROC curves were plotted (Fig. 2). The addition of plaque thickness alone to FRS resulted in significant increase in area under the curve (AUC) [from
0.60 (95% CI; 0.49e0.70) to 0.73 (95% CI; 0.63e0.82), p < 0.05]. Notably, the addition of plaque echogenicity (presence/absence of echolucent plaque) to the FRS and plaque thickness resulted in further and significant increase in AUC [from 0.73 (95% CI; 0.63e 0.82) to 0.82 (95% CI; 0.75e0.88), p < 0.05]. These results suggested that evaluation of the GSM, in addition to plaque thickness as well as conventional risk factors, can substantially improve the ability to predict the individuals with a high risk for CVD among asymptomatic type 2 diabetic patients. 4. Discussion Atherosclerotic changes such as intima-media thickness (IMT) and plaque formation in carotid artery, markers of early atherosclerosis and vascular remodeling that can be assessed quickly, non-invasively, and economically with high-resolution ultrasound, are correlated with conventional coronary risk factors and regarded as surrogates of CVD. Indeed, previous studies showed that carotid IMT was an independent predictor of CVD in asymptomatic patients [1,24e26]. However, previous studies also indicated that the addition of CIMT to conventional coronary risk factors could bring at most only a small improvement of the prediction ability for CVD [2,3,27]. It is commonly accepted that lipid-rich plaques carry a higher risk of CVD event, since “disruption of a lipid-rich vulnerable plaque” could play a crucial role in the development of CVD event [8,9]. Based on a working hypothesis that the individuals who have lipid-rich plaques in their carotid arteries are also prone to have
Y. Irie et al. / Atherosclerosis 230 (2013) 399e405
403
Table 2 Relative risk of cardiovascular events.
a
Age Male Smoking habits (0 ¼ no, 1 ¼ yes) Diabetes durationa BMIa HbA1ca Creatinine (mg/dL)a Hypertension Dyslipidemia Achievement of targeted values of blood pressure (0 ¼ yes, 1 ¼ no) Achievement of targeted values of LDL-C levels (0 ¼ yes, 1 ¼ no) FRSa Plaque number Plaque thicknessa Echolucent plaque (GSM 37)
Univariate model
Multivariate model
HR (95% CI)
HR (95% CI)
1.29 3.10 3.23 1.16 0.84 1.39 1.23 0.98 1.32 1.09
(0.91e1.84) (1.09e8.79)y (1.41e7.43)y (0.84e1.59) (0.59e1.20) (1.05e1.85)y (0.93e1.64) (0.41e2.37) (0.57e3.02) (0.55e2.13)
1.19 (0.92e1.56)
0.82 (0.41e1.62) 1.44 1.28 2.10 6.99
(1.07e1.94)y (1.09e1.50)z (1.71e3.28)z (3.46e14.14)z
Treatment approach for diabetes Diet only Insulin treatment Oral anti-hyperglycemic agents
2.41 (0.73e7.88) 1.38 (0.69e2.75) 0.76 (0.34e1.68)
Using antihypertensive drugs Using statins Using anti-aggregating agents
0.99 (0.49e2.00) 0.89 (0.44e1.80) 4.95 (2.52e9.72)z
1.30 1.10 1.44 4.55
(0.94e1.79) (0.91e1.32) (1.01e2.06)y (2.10e19.84)z
3.29 (1.64e6.58)z
Abbreviation: BMI, body mass index; LDL-C, low-density lipoprotein cholesterol; FRS, Framingham risk score; GSM, gray-scale median. HRs are given in bold. y p Value < 0.05. z p Value < 0.01. a Data are the HRs and 95% CIs in 1 SD increment.
Fig. 1. Association between the carotid plaque thickness, number of carotid plaques, and gray-scale median (GSM) and cardiovascular disease. Data are the HRs and 95% CIs of each quartile relative to the 1st quartile (Q1). (a) The quartiles of plaque thickness were as follows: Q1: 1.5 mm, Q2: 1.6e2.0 mm, Q3: 2.1e2.7 mm, and Q4: 2.8 mm. (b) The quartiles of plaque number were as follows: Q1: 1 or 2, Q2: 3, Q3: 4, and Q4: 5. (c) The quartiles of GSM were as follows: Q1: 59, Q2: 48e58, Q3: 38e47, and Q4: 37. yp Value < 0.05.
vulnerable plaques in their coronary and/or cerebral arteries, we prospectively evaluated whether ultrasonic tissue characterization of carotid plaque, together with plaque thickness, could improve the prediction ability of future CVD events in 287 type 2 diabetic patients without history of CVD. The current study first showed that patients with echolucent (GSM 37) plaque in the carotid artery had significantly higher risk for cardiovascular events (HR 6.99, 95% CI 3.46e14.14; p < 0.001). Furthermore, a multivariate Cox proportional hazards regression analysis revealed that the presence of echolucent carotid plaque was an independent predictor for CVD events, even after adjustment for plaque thickness as well as the Framingham risk score, which reflects conventional risk factors. These results were
consistent with the findings of previous studies that showed that echolucent plaque in the carotid artery corresponded to lipid-rich atheromatous plaque [17,28], and with the idea that patients with such vulnerable plaques in their coronary and/or cerebral arteries are prone to develop cardiovascular event. Although several studies have already shown that the presence of echolucent carotid plaque is associated with CVD events [21,29e34], this is the first study to show that the presence of echolucent carotid plaque identified by computer-assisted ultrasonic tissue characterization is an independent risk for CVD events in asymptomatic diabetic patients without apparent CVD.
Fig. 2. Time-dependent ROC curves for predicting cardiovascular disease (CVD). ROC curves were based on models of the prediction ability for CVD events with the use of FRS alone; FRS and plaque thickness; or FRS, plaque thickness, and plaque echogenicity (presence/absence of echolucent plaque). The AUCs for CVD were 0.60 (with FRS alone), 0.73 (with FRS and plaque thickness), and 0.82 (with FRS, plaque thickness, and plaque echogenicity). The addition of plaque echogenicity, together with plaque thicknesses, to FRS significantly increased the AUC (from 0.73 [0.63e0.82] to 0.82 [0.75e0.88]; p < 0.05).
404
Y. Irie et al. / Atherosclerosis 230 (2013) 399e405
Next, the prediction ability of low GSM and plaque thickness for identifying patients at high risk for CVD was then evaluated. The AUC estimated by time-dependent ROC curve analysis significantly increased after the addition of plaque thickness to FRS (from 0.60 to 0.73, p < 0.05). Interestingly, the AUC further increased after the addition of plaque echogenicity assessed with GSM to FRS and plaque thickness (from 0.73 to 0.82, p < 0.05). These findings suggested that the evaluation of the GSM of a carotid plaque, in addition to conventional measurement of its thickness and assessment of classical risk factors, could substantially improve the ability to predict the individuals with a high risk for CVD among asymptomatic type 2 diabetic patients. These results were compatible with a working hypothesis that assessment of tissue characteristics of carotid plaque as well as its size is useful for screening groups with a high risk of future cardiovascular event. To the best of our knowledge, this is the first report showing that the addition of GSM to conventional risk model improves the risk estimation for CVD event. Interestingly, in 184 (64%) out of 287 cases in this study, the thickest plaque also showed the lowest GSM value. Thus, when there were multiple carotid plaques in one individual, the thickest plaque was not always the plaque with the lowest GSM value. Therefore, it would be desirable to evaluate both the thickness and the echogenicity of each carotid plaque when assessing the risk for CVD. Several limitations of our study should be discussed. First, the number of study patients was relatively small, and there was a gender imbalance among participants. To conclusively demonstrate that the presence of echolucent carotid plaque is consistently and independently associated with cardiovascular events even after adjustment for conventional risk factors, many multivariate regression models should be tested. Unfortunately, such analyses were not able to be performed, since the number of patients who developed cardiovascular events during the observation period was not large enough. To compensate for this shortcoming, we adopted FRS as a risk prediction estimate that reflects accumulative effects of conventional risk factors. Although ROC curve analysis showed that the AUC of FRS for predicting risk of CVD in type 2 diabetic patients was only 0.60 in this study, this result was consistent with those of several previous studies, which also showed a value of around 0.60 [35,36]. Regardless, the clinical utility of ultrasonic tissue characterization of carotid plaque as a stratification tool for CVD risk should be verified in studies with larger sample size. Second, the precise cut-off level of GSM to detect individuals with high risk for CVD should be confirmed by further studies. We adopted the GSM values 0e5 for blood and 185e195 for adventitia, respectively, and found that the presence of GSM 37 plaque in the carotid was associated with higher incidence of CVD and that this cut-off level of GSM significantly improved the prediction ability for CVD. In similar settings, Falkowski et al. reported that plaques with GSM < 35 were associated with higher incidence of cerebral infarction [33], and Nicolaides et al. reported that cerebrovascular or retinal ischemic event occurred exclusively in patients with GSM < 30 [34]. Although these findings, including ours, indicated that GSM values in the range of 30e40 seem to be adequate cut-off values to detect individuals with high risk for CVD, this should be confirmed in further studies. Third, this study was limited to patients with carotid plaque. In the present study, 34 patients were not included in the analyses, because they had no carotid plaque. Interestingly, there were no incidences of CVD event in these 34 patients during the follow-up period, which suggests that patients without carotid plaque could be at low risk for CVD. Whether GSM values in carotid wall can provide any useful information for identifying individuals at high risk for CVD, even in patients without any carotid plaque, should be evaluated in another study.
Fourth, based on a cross-sectional analysis in 72 carotid endarterectomy specimens, Ciccone et al. reported that the degree of carotid plaque calcification was indicative of a high risk of coronary events [15], suggesting that the presence of calcified carotid plaques also serves as a marker of increased cardiovascular risk. Fifth, the ultrasound settings for each image were not always standardized. However, the blood was used as the reference for black and the adventitia as the reference for white, and gain settings for measurements within an individual were similar throughout the study. Therefore, the impact of gain of ultrasound beam on the GSM value would be quite small, if any. Sixth, the case of transient ischemic attack (n ¼ 1) should be excluded from the analyses, since it is not a hard event. However, a series of analyses showed similar results, regardless of whether this case was included. Seventh, the development of peripheral artery disease was considered as one of the components of primary outcome. However, in the present study, some asymptomatic peripheral artery disease cases might have been overlooked, since ankle-brachial indices were not measured in all the study patients. Finally, the present study focused on the question of whether measurement of GSM values can provide any useful information for identifying patients with a high risk for CVD, event at a certain point in their life with diabetes. Our thesis would have been strengthened if it were shown that measurement of GSM could improve the risk prediction of cardiovascular event after adjustment for the changes in cardio-metabolic control and therapeutic regimen during the follow-up period. However, we were not able to include many parameters into the multivariate regression model, because the number of patients in this study was not large enough. Whether longitudinal monitoring of these parameters during the follow-up period affects the prediction ability of GSM should be evaluated in further studies. Notwithstanding these limitations, our study indicates that measurement of carotid GSM can provide useful information for identifying patients with a high risk of CVD. In conclusion, the present study suggests that ultrasonic tissue characterization of carotid plaque using the GSM, a quantitative parameter of the plaque echogenicity, can improve the risk prediction of cardiovascular event in asymptomatic type 2 diabetic patients with carotid plaque. Author contributions Y.I., N.K., M.T. and H.K. researched data and wrote the manuscript. M.N., R.K., K.S., Y.Um., and S.S. researched the data. Y.Ue., K.K, and I.S. contributed to the interpretation of the results and the discussion. All authors reviewed and approved the report. N.K. is the guarantor of this work and, as such, had full access to all of the data in the study and takes responsibility for the integrity of the data and accuracy of the data analysis. Funding sources The Diabetes Masters Conference supported by the Japan Diabetes Foundation and Grants-in-Aid for Scientific Research from the Japanese Ministry of Education, Science, Sports, Culture and Technology [KAKENHI 25461349]. Conflict of interest disclosures None.
Y. Irie et al. / Atherosclerosis 230 (2013) 399e405
Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.atherosclerosis.2013.08.015. References [1] O’Leary DH, Polak JF, Kronmal RA, Manolio TA, Burke GL, Wolfson Jr SK. Carotid-artery intima and media thickness as a risk factor for myocardial infarction and stroke in older adults. Cardiovascular Health Study Collaborative Research Group. N Engl J Med 1999;340:14e22. [2] Nambi V, Chambless L, Folsom AR, et al. Carotid intima-media thickness and presence or absence of plaque improves prediction of coronary heart disease risk: the ARIC (Atherosclerosis Risk in Communities) study. J Am Coll Cardiol 2010;55:1600e7. [3] Polak JF, Pencina MJ, Pencina KM, O’Donnell CJ, Wolf PA, D’Agostino Sr RB. Carotid-wall intima-media thickness and cardiovascular events. N Engl J Med 2011;365:213e21. [4] Mathiesen EB, Johnsen SH, Wilsgaard T, Bønaa KH, Løchen ML, Njølstad I. Carotid plaque area and intima-media thickness in prediction of first-ever ischemic stroke: a 10-year follow-up of 6584 men and women: the Tromsø study. Stroke 2011;42:972e8. [5] Johnsen SH, Mathiesen EB, Joakimsen O, et al. Carotid atherosclerosis is a stronger predictor of myocardial infarction in women than in men: a 6-year follow-up study of 6226 persons: the Tromsø study. Stroke 2007;38:2873e80. [6] Fuster V, Badimon L, Badimon JJ, Chesebro JH. The pathogenesis of coronary artery disease and the acute coronary syndromes. N Engl J Med 1992;326:242e50. [7] Lee RT, Grodzinsky AJ, Frank EH, Kamm RD, Schoen FJ. Structure-dependent dynamic mechanical behavior of fibrous caps from human atherosclerotic plaques. Circulation 1991;83:1764e70. [8] Davies MJ, Richardson PD, Woolf N, Katz DR, Mann J. Risk of thrombosis in human atherosclerotic plaques: role of extracellular lipid, macrophage, and smooth muscle cell content. Br Heart J 1993;69:377e81. [9] Falk E. Pathogenesis of atherosclerosis. J Am Coll Cardiol 2006;47:C7e12. [10] Nishio M, Ueda Y, Matsuo K, et al. Detection of disrupted plaques by coronary CT: comparison with angioscopy. Heart 2011;97:1397e402. [11] Tahara N, Kai H, Nakaura H, et al. The prevalence of inflammation in carotid atherosclerosis: analysis with fluorodeoxyglucose-positron emission tomography. Eur Heart J 2007;28:2243e8. [12] Hatsukami TS, Ross R, Polissar NL, Yuan C. Visualization of fibrous cap thickness and rupture in human atherosclerotic carotid plaque in vivo with high-resolution magnetic resonance imaging. Circulation 2000;102:959e64. [13] General findings of the international atherosclerosis project. Lab Invest 1968;18:498e502. [14] Dalager S, Paaske WP, Kristensen IB, Laurberg JM, Falk E. Artery-related differences in atherosclerosis expression: implications for atherogenesis and dynamics in intima-media thickness. Stroke 2007;38:2698e705. [15] Ciccone MM, Marzullo A, Mizio D, et al. Can carotid plaque histology selectively predict the risk of an acute coronary syndrome? Int Heart J 2011;52:72e7. [16] Katakami N, Takahara M, Kaneto H, et al. Ultrasonic tissue characterization of carotid plaque improves the prediction of cardiovascular events in diabetic patients: a pilot study. Diabetes Care 2012;35:2640e6. [17] Grønholdt ML, Wiebe BM, Laursen H, Nielsen TG, Schroeder TV, Sillesen H. Lipid-rich carotid artery plaques appear echolucent on ultrasound B-mode images and may be associated with intraplaque haemorrhage. Eur J Vasc Endovasc Surg 1997;14:439e45. [18] Japan Diabetes Society. Diabetes mellitus treatment guideline based on scientific ground. Revised 2nd ed. Nankodo: Japan Diabetes Society; 2007. p. 257e72.
405
[19] D’Agostino Sr RB, Vasan RS, Pencina MJ, et al. General cardiovascular risk profile for use in primary care: the Framingham heart study. Circulation 2008;117:743e53. [20] Terminology and Diagnostic Criteria Committee, Japan Society of Ultrasonics in Medicine, Subcommittee for Preparing Guidelines for Ultrasound Diagnosis of Carotid Artery. Standard method for ultrasound evaluation of carotid artery lesions. Jpn J Med Ultrasonics 2009;36:501e18. [21] Sabetai MM, Tegos TJ, Nicolaides AN, Dhanjil S, Pare GJ, Stevens JM. Reproducibility of computer-quantified carotid plaque echogenicity: can we overcome the subjectivity? Stroke 2000;31:2189e96. [22] Smith Jr SC, Dove JT, Jacobs AK, et al., American College of Cardiology; American Heart Association Task Force on Practice Guidelines. Committee to Revise the 1993 Guidelines for Percutaneous Transluminal Coronary Angioplasty. ACC/AHA guidelines of percutaneous coronary interventions (revision of the 1993 PTCA guidelines) e executive summary. A report of the American College of Cardiology/American Heart Association task force on practice guidelines (committee to revise the 1993 guidelines for percutaneous transluminal coronary angioplasty). J Am Coll Cardiol 2001;37:2215e39. [23] Heagerty PJ, Lumley T, Pepe MS. Time-dependent ROC curves for censored survival data and a diagnostic marker. Biometrics 2000;56:337e44. [24] Salonen JT, Salonen R. Ultrasonographically assessed carotid morphology and the risk of coronary heart disease. Arterioscler Thromb 1991;11:1245e9. [25] Bots ML, Hoes AW, Koudstaal PJ, Hofman A, Grobbee DE. Common carotid intima-media thickness and risk of stroke and myocardial infarction: the Rotterdam study. Circulation 1997;96:1432e7. [26] Chambless LE, Folsom AR, Clrgg LX. Carotid wall thickness is predictive of incident clinical stroke: the Atherosclerosis Risk in Communities (ARIC) study. Am J Epidemiol 2000;151:478e87. [27] del Sol AI, Moons KG, Hollander M, et al. Is carotid intima-media thickness useful in cardiovascular disease risk assessment? The Rotterdam study. Stroke 2001;32:1532e8. [28] Lal BK, Hobson 2nd RW, Pappas PJ, et al. Pixel distribution analysis of B-mode ultrasound scan images predicts histologic features of atherosclerotic carotid plaques. J Vasc Surg 2002;35:1210e7. [29] Mathiesen EB, Bønaa KH, Joakimsen O. Echolucent plaques are associated with high risk of ischemic cerebrovascular events in carotid stenosis: the Tromsø study. Circulation 2001;103:2171e5. [30] Honda O, Sugiyama S, Kugiyama K, et al. Echolucent carotid plaques predict future coronary events in patients with coronary artery disease. J Am Coll Cardiol 2004;43:1177e84. [31] Seo Y, Watanabe S, Ishizu T, et al. Echolucent carotid plaques as a feature in patients with acute coronary syndrome. Circ J 2006;70:1629e34. [32] Grønholdt ML, Nordestgaard BG, Schroeder TV, Vorstrup S, Sillesen H. Ultrasonic echolucent carotid plaques predict future strokes. Circulation 2001;104: 68e73. [33] Falkowski A, Kaczmarczyk M, Cieszanowski A, Goracy I, Poncyliusz W, Wilk G. Computer-assisted characterisation of a carotid plaque. Med Sci Monit 2004;10:67e70. [34] Nicolaides AN, Kakkos SK, Kyriacou E, et al., Asymptomatic Carotid Stenosis and Risk of Stroke (ACSRS) Study Group. Asymptomatic internal carotid artery stenosis and cerebrovascular risk stratification. J Vasc Surg 2010;52:1486e96 e1e5. [35] van Dieren S, Peelen LM, Nöthlings U, et al. External validation of the UK prospective diabetes study (UKPDS) risk engine in patients with type 2 diabetes. Diabetologia 2011;54:264e70. [36] Kengne AP, Patel A, Colagiuri S, et al., ADVANCE Collaborative Group. The Framingham and UK prospective diabetes study (UKPDS) risk equations do not reliably estimate the probability of cardiovascular events in a large ethnically diverse sample of patients with diabetes: the action in diabetes and vascular disease: preterax and diamicron-MR controlled evaluation (ADVANCE) study. Diabetologia 2010;53:821e31.
Contents lists available at ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
The utility of ultrasonic tissue characterization of carotid plaque in the prediction of cardiovascular events in diabetic patients Yoko Irie a, Naoto Katakami a, b, *, Hideaki Kaneto a, Mitsuyoshi Takahara a, Mayu Nishio c, Ryuichi Kasami a, Ken’ya Sakamoto c, Yutaka Umayahara d, Satoru Sumitsuji e, Yasunori Ueda c, Keisuke Kosugi c, Iichiro Shimomura a a
Department of Metabolic Medicine, Osaka University Graduate School of Medicine, Japan Department of Metabolism and Atherosclerosis, Osaka University Graduate School of Medicine, Japan Osaka Police Hospital, Japan d Osaka General Medical Center, Japan e Department of Advanced Cardiovascular Therapeutics/Cardiology, Osaka University Graduate School of Medicine, Japan b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 April 2013 Received in revised form 15 August 2013 Accepted 15 August 2013 Available online 26 August 2013
Objective: The aim of this study was to evaluate whether non-invasive ultrasonic tissue characterization of carotid plaque using gray-scale median (GSM) can be a predictor of future cardiovascular disease (CVD) events in type 2 diabetic patients. Methods: A total of 287 type 2 diabetic patients with carotid plaque but without CVD were enrolled (male 72%, mean age 65 7 years). We prospectively evaluated the association between GSM, a quantitative parameter of the plaque echogenicity, and CVD. Results: The median follow-up period was 55 months, and there were 34 new CVD events. The risk of CVD event was significantly higher in the patients with echolucent (GSM 37) plaque (n ¼ 67) as compared to those without (n ¼ 220) (HR ¼ 6.99, 95% CI 3.46e14.14, p < 0.001). Cox proportional hazards regression analysis showed that the presence of echolucent plaque (HR ¼ 4.55, 95% CI 2.10e19.84, p < 0.001) as well as plaque thickness (HR ¼ 1.44, 95% CI 1.01e2.06, p ¼ 0.005) were independent predictors of CVD, even after adjustment for other risk factors. Time-dependent receiver-operatingcharacteristic curve analysis revealed that the addition of plaque thickness to Framingham risk score (FRS) resulted in significant increase in area under the curve (AUC) [from 0.60 (95% CI; 0.49e0.70) to 0.73 (95% CI; 0.63e0.82), p < 0.05]. Notably, the addition of plaque echogenicity (presence/absence of echolucent plaque) to the FRS and plaque thickness resulted in further and significant increase in AUC [from 0.73 (95% CI; 0.63e0.82) to 0.82 (95% CI; 0.75e0.88), p < 0.05]. Conclusion: Ultrasonic tissue characterization of carotid plaque using the GSM can improve the risk prediction of cardiovascular event in asymptomatic type 2 diabetic patients with carotid plaque. Ó 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Carotid arteries Cardiovascular diseases Diabetes mellitus Vulnerable plaque Carotid ultrasound
1. Introduction Since cardiovascular disease (CVD) remains the main cause of death and impairment of quality of life in diabetic patients, early identification of individuals at high risk for CVD events and subsequent rapid interventions are important. Recently, ultrasound examination of the carotid artery has been in use to non-invasively
* Corresponding author. Department of Metabolic Medicine, Osaka University Graduate School of Medicine, 2-2, Yamadaoka, Suita, Osaka 565-0871, Japan. Tel.: þ81 6 6879 3743; fax: þ81 6 6879 3739. E-mail address: [email protected] (N. Katakami). 0021-9150/$ e see front matter Ó 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atherosclerosis.2013.08.015
identify individuals at high risk for CVD. Some studies have shown that the presence/absence of carotid plaque assessed by ultrasound could improve the prediction ability for CVD over and above conventional coronary risk factors [1e5]. It is well known that disruption of an atherosclerotic plaque plays a crucial role in the pathogenesis of CVD event and that plaque disruption is dependent on the content of lipid, neovascular vessel, and inflammatory cells in the atheroma, and the thickness of the fibrous cap [6e9]. Therefore, tissue characterization of a plaque lesion is considered to be useful for identifying patients at high risk for CVD and death. Presently, various modalities such as Computed Tomography (CT), Magnetic Resonance Imaging (MRI), or Fluorodeoxyglucose-Positron Emission Tomography (FDG-PET) are
400
Y. Irie et al. / Atherosclerosis 230 (2013) 399e405
being used to detect vulnerable plaque in coronary and carotid arteries [10e12]. However, it is unrealistic to screen all the asymptomatic patients for individuals at high risk for CVD with these tools, since these tests are limited by the potential of significant adverse effects, technical difficulty, availability, and cost. Carotid arteries as well as the aorta develop foam cell lesions and lipid core plaque at an early age [13]. Interestingly, a recent study reported that the prevalence of lipid-rich plaques in carotid arteries was higher in coronary death than in non-coronary death [14]. Another study using carotid endarterectomy specimens also reported that a soft carotid plaque with a large necrotic core and a marked inflammatory component may be predictive of an increased risk of cerebrovascular events [15]. These findings indicate that the individuals who have vulnerable plaques in their carotid arteries are prone to also have vulnerable plaques in their coronary and/or cerebral arteries, which will lead to CVD events, and thus, the assessment of tissue characteristics of carotid plaques will help to identify the patients with a high risk for CVD event. Indeed, based on a prospective study in 85 asymptomatic type 2 diabetic patients, we have already shown that a quantitative ultrasonic tissue characterization of carotid plaque using integrated backscatter (IBS) analysis can be a predictor of future CVD event [16]. However, the measurement of IBS values requires a specific software package, “Acoustic Densitometry,” installed in a SONOS echocardiograph system (Philips Medical Systems), which would limit the widespread use of this approach. In a qualitative evaluation using conventional B-mode ultrasound imaging performed by experienced sonographers, vulnerable plaques, which consist mainly of neovascular vessel, high lipid content, and inflammatory infiltration [8,9], are believed to appear “hypoechoic.” On the other hand, stable plaques, which consist mainly of fibrous tissue and calcific components, appear “hyperechoic” [17]. Recently, echogenicity of carotid plaques has become semi-quantitatively evaluable with high-resolution B-mode ultrasound and computer-assisted image processing: the gray-scale median (GSM) of the frequency distribution of gray values of the pixels within the plaque serves as a measure of echogenicity. Since the evaluation of GSM values in carotid plaques requires neither specific software nor a specific ultrasonograph, it can be a more universal and practical approach as compared with IBS analysis. Based on the background described above, the current study prospectively evaluated whether non-invasive and inexpensive ultrasonic tissue characterization of carotid plaque using GSM can improve the risk prediction of CVD event in asymptomatic type 2 diabetic patients with carotid plaque. 2. Research design and methods 2.1. Subjects Middle-aged and older Japanese type 2 diabetic patients with carotid plaque but without apparent CVD participated in this study. Patients who fulfilled the following criteria were considered eligible: (1) age 40 years at the time of enrollment, and (2) diagnosed with type 2 diabetes based on the criteria of the Japan Diabetes Society. Exclusion criteria were (1) history of ischemic stroke, coronary heart disease, or peripheral artery disease; (2) elevated liver enzymes (GOT or GPT 2.5 times the normal range); and (3) renal insufficiency (serum creatinine 2.0 mg/dL). Screening of the study patients was performed consecutively during the registration period (from April 2007 to December 2009) at the outpatient diabetes clinic of Osaka Police Hospital. All the patients that met eligibility criteria were asked whether they could participate in the present study, and all the patients who agreed to participate were registered. Out of 325 patients who had been
registered, four were excluded because of poor image or acoustic shadowing due to severe calcification, and 34 were excluded because of no plaque. After the exclusions, a total of 287 patients were analyzed in this study. This study was conducted in agreement with the principles of the Helsinki declaration. The study protocol was approved by the committee on human research ethics of Osaka Police Hospital. Written informed consent was obtained from all the participants after a full explanation of the study. 2.2. Clinical and biochemical analyses Blood samples were collected after an overnight fast for analysis of serum concentrations of glucose, total cholesterol, triglycerides, HDL cholesterol, LDL cholesterol, and HbA1c. Diagnosis of type 2 diabetes was made by the principal investigator in accordance with the standards of the Japan Diabetes Society [18]. Dyslipidemia was defined as LDL cholesterol 120 mg/dL, HDL cholesterol < 40 mg/ dL, triglycerides 150 mg/dL, or current use of lipid-lowering agents in accordance with the standards of the Japan Diabetes Society [18]. Hypertension was defined as systolic blood pressure 130 mmHg and/or diastolic blood pressure 80 mmHg or current use of antihypertensive agents. The risk of CVD was estimated using the Framingham D’Agostino equations [19]. Risk factors included in the model to assess risk of CVD were gender, age, total cholesterol, HDL cholesterol, systolic blood pressure, antihypertensive medication use, current smoking, and diabetes status. 2.3. Ultrasound examination B-mode ultrasonography of the carotid artery was performed using an ultrasound machine (Toshiba SSA-790CE; Toshiba Medical Systems, Tokyo, Japan) with a 7.5-MHz linear transducer. All scanning was conducted by experienced laboratory physicians using the same ultrasound system and the same measuring method. Scanning of the extracranial common carotid artery (CCA), the carotid bulb (CB), and the internal carotid artery (ICA) in the neck was performed bilaterally in three different longitudinal projections and three different transverse projections. According to the guideline of the Japan Society of Ultrasonics [20], carotid plaque was defined as a focal structure encroaching into the arterial lumen or demonstrating a thickness > 1.0 mm as measured from the mediaeadventitia interface to the intra-lumen interface. The plaque thickness in the CCA, the CB, and the ICA were measured separately, and the greatest value among them was used as the representative value for each individual. Reproducibility analysis of 20 replicate measurements yielded absolute mean differences of 0.03 0.05 mm for plaque thickness. The inter-observer coefficient of variation for measurement of plaque thickness was 0.7%. The plaque echogenicity was evaluated based on the gray-scale median (GSM) in a gray-scale range of 0e255 (0 as the darkest and 255 as the brightest tone). Adobe Photoshop software (Adobe Systems, version 7.0, San Jose, CA, USA) was used for image standardization and calculation of gray-scale values. According to the criteria reported previously, the standardization of the B-mode image was performed by using a curve option, so that the GSM for the blood ranged from 0 to 5 and for the adventitia from 185 to 195 [21]. Thus the gray-scale values of all pixels would change according to the new linear scale defined by the reference values for blood and adventitia. The plaque was then delineated with a freehand tool, and the GSM of each plaque was read from the entire delineated area. In case there were multiple plaques in one individual, all the plaques present were subject to GSM measurement and the lowest value among them was used as the representative
Y. Irie et al. / Atherosclerosis 230 (2013) 399e405
value for the patients. To avoid inter-reader variability, all scans were electronically stored, sent to a central office, and read in a random order by a single reader (Y. I.) unaware of the clinical characteristics of the patients. The intra-reader coefficient of variation for GSM measurements was 5.8% for 140 consecutively replicated measurements (male 70%, age 67 7.3 years). 2.4. Assessment of cardiovascular events The prespecified primary endpoint was the first occurrence of fatal or non-fatal CVD during the follow-up period, which was a composite of any coronary heart disease (angina [both unstable and stable], myocardial infarction [both ST elevation myocardial infarction (STEMI) and non-ST elevation myocardial infarction (NSTEMI)], and coronary revascularization), any ischemic stroke (transient ischemic attack, fatal and non-fatal ischemic stroke), and peripheral artery disease during the follow-up period. The diagnosis of the occurrence of coronary heart disease was performed by cardiologists based on the clinical symptoms, characteristic electrocardiogram changes, cardiac enzyme levels, and the findings from coronary angiography and/or echocardiography, according to established guidelines. Indication for coronary revascularization for patients with stable coronary artery disease was decided based on ACC/AHA guidelines for percutaneous coronary intervention [22]. An ischemic stroke event was defined as a validated definite or probable hospitalized atherothrombotic, cardioembolic, lacunar, or other type of ischemic stroke diagnosed by neurosurgical experts based on clinical symptoms and neuroimaging findings, according to the National Institute of Neurologic Disorders and Stroke III classification. Peripheral artery disease was defined as patients with intermittent claudication due to atherosclerosis of the lower extremity arteries; evaluated based on ankle-brachial index. For participants with incident CVD event, follow-up was defined as the period between the baseline clinic visit and the date of the first event. For participants with no CVD event, follow-up continued until the date of death or December 11th 2012 or until the date of last contact. All the study patients were recommended to visit Osaka Police Hospital for treatment every 1e2 months throughout the follow-up period. At each hospital visit, medical interview, physical examination, and blood examination were performed, and the occurrence of CVD events was then determined. For a potential new CVD event, even if it was recognized at another medical institution, additional information, including the results of imaging and other diagnostic procedures, was obtained for confirmation, and all causes of death were confirmed by medical certificate. 2.5. Statistical analyses All statistical analyses were performed with SPSS version 11.0J for Windows (SPSS, Inc., Japan). All values are reported as mean SD, median (IQR), or real number of patients with the percentage in parentheses. Cox proportional hazards regression model was used to determine the unadjusted association of each variable with the outcome and hazard ratios (HRs) and 95% CIs were reported. The statistically significant variables in the unadjusted analyses were entered into multivariate models to reveal the independent impact on the outcome. The prediction ability of variables to predict CVD was examined by time-dependent receiver-operating-characteristic (ROC) curve analysis [23]. Curves were generated based on models of the prediction of risk with the use of Framingham risk score (FRS) alone, or with plaque thickness, or with plaque thickness and plaque echogenicity (presence/ absence of echolucent plaque). For all tests, p < 0.05 was considered statistically significant.
401
3. Results 3.1. Patient characteristics The baseline characteristics of the study population are shown in Table 1. Among a total of 287 patients (males, 72%; age, 65 7 years (mean SD); diabetes duration, 13 9 years; HbA1c, 7.5 1.1%.), 236 (82%) patients had hypertension and 215 (75%) had dyslipidemia. During a median follow-up of 55 months, 34 of the 287 patients (12%) had new CVD events: 11 had ischemic stroke, 21 had coronary heart disease, and 2 had peripheral artery disease (Supplementary Table 1). During the observation period, 4 participants died (2 were cardiac sudden deaths and the other 2 were cancer deaths). As compared to the patients who did not develop CVD during the observation period, the patients who developed CVD were significantly more likely to be male, and showed significantly higher smoking rates and HbA1c values (Table 1). There was no significant difference in the baseline therapeutic regimens for diabetes, dyslipidemia, or hypertension between the patients who developed CVD and those who did not. The baseline prescription rate of anti-aggregating agents was significantly greater in patients with CVD events. 3.2. Association between plaque echogenicity and cardiovascular events in patients with type 2 diabetes mellitus At baseline, plaque thickness was significantly greater (2.98 1.05 mm with CVD events vs. 2.16 0.79 mm without, p < 0.001) and plaque number was significantly greater (4.4 2.0 with CVD events vs. 3.4 1.8 without, p ¼ 0.005) in patients who developed CVD events during the observation period as compared to those who did not (Table 1). The risks for CVD events were also compared among the quartiles of plaque thickness and plaque number, and it was confirmed that both plaque thickness and plaque number showed a positive relationship with the risk of CVD events. Compared to the lowest quartile of plaque thickness (1.5 mm), the highest quartile (2.8 mm) had a significantly increased risk. Compared to the lowest quartile of plaque number (plaque number ¼ 1 or 2), the 3rd (4) and the highest quartile (5) had a significantly increased risk (Fig. 1a and b). GSM value was significantly lower in patients with CVD events as compared to those without (38 16 with CVD events vs. 51 17 without, p < 0.001). In 184 (64%) out of 287 cases, the thickest plaque also showed the lowest GSM value. First, to evaluate the relationship between the echogenicity in carotid plaque and the occurrence of CVD events in detail, the GSM dataset was divided into quartiles (Q1: 59, Q2: 48e58, Q3: 38e47, and Q4: 37) and the decrement of GSM level and the risk for CVD events were analyzed. The HRs and 95% CIs of Q2, Q3, and Q4 relative to Q1 were 0.54 (95% CI 0.13e2.24; p ¼ 0.393), 0.75 (95% CI 0.20e2.81; p ¼ 0.674), and 5.28 (95% CI 2.00e13.94; p ¼ 0.001), respectively (Fig. 1c). Thus, the lowest GSM quartile (Q4: GSM 37) had significantly and much higher risk for CVD as compared to the other GSM quartiles, and the association between GSM values and the risk of CVD was not linear. Therefore, the subsequent statistical analyses were performed based on the hypothesis that carotid plaques with GSM value 37 are considered to be clinically “echolucent” plaques. When the study patients were divided into two groups (patients with low (37) GSM plaques (n ¼ 67) and those with high (>37) GSM plaques (n ¼ 220)), the risk of cardiovascular event was significantly higher in the former as compared to the latter (HR 6.99, 95% CI 3.46e14.14; p < 0.001). Similarly, patients with echolucent (GSM 37) plaque had significantly higher risk for ischemic stroke (HR 5.86, 95% CI 1.72e20.00;
402
Y. Irie et al. / Atherosclerosis 230 (2013) 399e405
Table 1 Baseline characteristics of the study population. All subjects (n ¼ 287)
Cardiovascular events
p Value
(þ) (n ¼ 34)
() (n ¼ 253)
Age (years) Male Smoking habits (yes/no) Diabetes duration (years) BMI (kg/m2) HbA1c (%) Creatinine (mg/dL) Hypertension Dyslipidemia Plaque number Plaque thickness (mm) GSM
65 7 207 (72) 161 (56) 10 (7e17) 24 3 7.5 1.1 0.9 0.2 236 (82) 215 (75) 3.5 1.9 2.26 0.86 49 17
67 8 30 (88) 27 (79) 13 (7e21) 24 3 7.9 1.3 0.9 0.3 28 (82) 27 (79) 4.4 2.0 2.98 1.05 38 16
65 7 177 (70) 134 (53) 10 (7e17) 24 3 7.5 1.1 0.8 0.2 208 (82) 188 (74) 3.4 1.8 2.16 0.79 51 17
NS 0.025 0.005 NS NS 0.016 NS NS NS 0.005 <0.001 <0.001
Treatment approach for diabetes Diet only Insulin treatment Oral anti-hyperglycemic agents
13 (5) 89 (31) 232 (81)
3 (9) 13 (38) 26 (76)
10 (4) 76 (30) 206 (81)
NS NS NS
Achievement of targeted values of blood pressure
129 (45)
16 (47)
113 (45)
NS
Treatment approach for hypertension Using antihypertensive drugs Calcium-channel blockers ACEI or ARB
187 (65) 81 (28) 150 (52)
22 (65) 8 (24) 19 (56)
165 (65) 73 (29) 131 (52)
NS NS NS
Achievement of targeted values of LDL-C levels
179 (62)
20 (59)
159 (63)
NS
Treatment approach for dyslipidemia Statins
108 (38)
12 (35)
96 (38)
NS
Using anti-aggregating agents
60 (21)
18 (53)
42 (17)
<0.001
Abbreviation: BMI, body mass index; GSM, gray-scale median; LDL-C, low-density lipoprotein cholesterol; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker. Based on the guideline of the Japan Diabetes Society, the targeted values of blood pressure and LDL-C levels were set at <130/80 mmHg and <120 mg/dL, respectively. All values are reported as mean SD, median (IQR), or real number of subjects with the percentage in parentheses. p Values compared with diabetes patients with and without cardiovascular events by unpaired-t, ManneWhitney U, or Chi square test.
p ¼ 0.005) and coronary heart disease (HR 7.13, 95% CI 2.88e17.68; p < 0.001), respectively. Next, a multivariate Cox proportional hazard regression model was used, in which HbA1c, Framingham risk score (FRS), plaque thickness, plaque number, the use of anti-aggregating drugs, and the presence of echolucent plaque were included as independent variables. This analysis revealed that the presence of echolucent plaque (HR 4.55 [95% CI 2.10e19.84]; p < 0.001) as well as plaque thickness (HR 1.44 [1.01e2.06]; p ¼ 0.005) were independently associated with CVD, even after adjustment for other factors (Table 2). To confirm that the presence of echolucent plaque was an independent predictor of CVD events even after adjustment for renal function and the treatment regimen, other multivariate Cox proportional hazard regression models were also tested. These analyses showed that the presence of echolucent plaque was an independent predictor of CVD events, even after adjustment for serum creatinine levels and the treatment regimen at baseline (data not shown). 3.3. Contribution of GSM in the prediction of cardiovascular event in patients with type 2 diabetes mellitus To examine whether the addition of plaque echogenicity and/or plaque thickness to conventional risk factors could improve the prediction ability for CVD event, time-dependent ROC curves were plotted (Fig. 2). The addition of plaque thickness alone to FRS resulted in significant increase in area under the curve (AUC) [from
0.60 (95% CI; 0.49e0.70) to 0.73 (95% CI; 0.63e0.82), p < 0.05]. Notably, the addition of plaque echogenicity (presence/absence of echolucent plaque) to the FRS and plaque thickness resulted in further and significant increase in AUC [from 0.73 (95% CI; 0.63e 0.82) to 0.82 (95% CI; 0.75e0.88), p < 0.05]. These results suggested that evaluation of the GSM, in addition to plaque thickness as well as conventional risk factors, can substantially improve the ability to predict the individuals with a high risk for CVD among asymptomatic type 2 diabetic patients. 4. Discussion Atherosclerotic changes such as intima-media thickness (IMT) and plaque formation in carotid artery, markers of early atherosclerosis and vascular remodeling that can be assessed quickly, non-invasively, and economically with high-resolution ultrasound, are correlated with conventional coronary risk factors and regarded as surrogates of CVD. Indeed, previous studies showed that carotid IMT was an independent predictor of CVD in asymptomatic patients [1,24e26]. However, previous studies also indicated that the addition of CIMT to conventional coronary risk factors could bring at most only a small improvement of the prediction ability for CVD [2,3,27]. It is commonly accepted that lipid-rich plaques carry a higher risk of CVD event, since “disruption of a lipid-rich vulnerable plaque” could play a crucial role in the development of CVD event [8,9]. Based on a working hypothesis that the individuals who have lipid-rich plaques in their carotid arteries are also prone to have
Y. Irie et al. / Atherosclerosis 230 (2013) 399e405
403
Table 2 Relative risk of cardiovascular events.
a
Age Male Smoking habits (0 ¼ no, 1 ¼ yes) Diabetes durationa BMIa HbA1ca Creatinine (mg/dL)a Hypertension Dyslipidemia Achievement of targeted values of blood pressure (0 ¼ yes, 1 ¼ no) Achievement of targeted values of LDL-C levels (0 ¼ yes, 1 ¼ no) FRSa Plaque number Plaque thicknessa Echolucent plaque (GSM 37)
Univariate model
Multivariate model
HR (95% CI)
HR (95% CI)
1.29 3.10 3.23 1.16 0.84 1.39 1.23 0.98 1.32 1.09
(0.91e1.84) (1.09e8.79)y (1.41e7.43)y (0.84e1.59) (0.59e1.20) (1.05e1.85)y (0.93e1.64) (0.41e2.37) (0.57e3.02) (0.55e2.13)
1.19 (0.92e1.56)
0.82 (0.41e1.62) 1.44 1.28 2.10 6.99
(1.07e1.94)y (1.09e1.50)z (1.71e3.28)z (3.46e14.14)z
Treatment approach for diabetes Diet only Insulin treatment Oral anti-hyperglycemic agents
2.41 (0.73e7.88) 1.38 (0.69e2.75) 0.76 (0.34e1.68)
Using antihypertensive drugs Using statins Using anti-aggregating agents
0.99 (0.49e2.00) 0.89 (0.44e1.80) 4.95 (2.52e9.72)z
1.30 1.10 1.44 4.55
(0.94e1.79) (0.91e1.32) (1.01e2.06)y (2.10e19.84)z
3.29 (1.64e6.58)z
Abbreviation: BMI, body mass index; LDL-C, low-density lipoprotein cholesterol; FRS, Framingham risk score; GSM, gray-scale median. HRs are given in bold. y p Value < 0.05. z p Value < 0.01. a Data are the HRs and 95% CIs in 1 SD increment.
Fig. 1. Association between the carotid plaque thickness, number of carotid plaques, and gray-scale median (GSM) and cardiovascular disease. Data are the HRs and 95% CIs of each quartile relative to the 1st quartile (Q1). (a) The quartiles of plaque thickness were as follows: Q1: 1.5 mm, Q2: 1.6e2.0 mm, Q3: 2.1e2.7 mm, and Q4: 2.8 mm. (b) The quartiles of plaque number were as follows: Q1: 1 or 2, Q2: 3, Q3: 4, and Q4: 5. (c) The quartiles of GSM were as follows: Q1: 59, Q2: 48e58, Q3: 38e47, and Q4: 37. yp Value < 0.05.
vulnerable plaques in their coronary and/or cerebral arteries, we prospectively evaluated whether ultrasonic tissue characterization of carotid plaque, together with plaque thickness, could improve the prediction ability of future CVD events in 287 type 2 diabetic patients without history of CVD. The current study first showed that patients with echolucent (GSM 37) plaque in the carotid artery had significantly higher risk for cardiovascular events (HR 6.99, 95% CI 3.46e14.14; p < 0.001). Furthermore, a multivariate Cox proportional hazards regression analysis revealed that the presence of echolucent carotid plaque was an independent predictor for CVD events, even after adjustment for plaque thickness as well as the Framingham risk score, which reflects conventional risk factors. These results were
consistent with the findings of previous studies that showed that echolucent plaque in the carotid artery corresponded to lipid-rich atheromatous plaque [17,28], and with the idea that patients with such vulnerable plaques in their coronary and/or cerebral arteries are prone to develop cardiovascular event. Although several studies have already shown that the presence of echolucent carotid plaque is associated with CVD events [21,29e34], this is the first study to show that the presence of echolucent carotid plaque identified by computer-assisted ultrasonic tissue characterization is an independent risk for CVD events in asymptomatic diabetic patients without apparent CVD.
Fig. 2. Time-dependent ROC curves for predicting cardiovascular disease (CVD). ROC curves were based on models of the prediction ability for CVD events with the use of FRS alone; FRS and plaque thickness; or FRS, plaque thickness, and plaque echogenicity (presence/absence of echolucent plaque). The AUCs for CVD were 0.60 (with FRS alone), 0.73 (with FRS and plaque thickness), and 0.82 (with FRS, plaque thickness, and plaque echogenicity). The addition of plaque echogenicity, together with plaque thicknesses, to FRS significantly increased the AUC (from 0.73 [0.63e0.82] to 0.82 [0.75e0.88]; p < 0.05).
404
Y. Irie et al. / Atherosclerosis 230 (2013) 399e405
Next, the prediction ability of low GSM and plaque thickness for identifying patients at high risk for CVD was then evaluated. The AUC estimated by time-dependent ROC curve analysis significantly increased after the addition of plaque thickness to FRS (from 0.60 to 0.73, p < 0.05). Interestingly, the AUC further increased after the addition of plaque echogenicity assessed with GSM to FRS and plaque thickness (from 0.73 to 0.82, p < 0.05). These findings suggested that the evaluation of the GSM of a carotid plaque, in addition to conventional measurement of its thickness and assessment of classical risk factors, could substantially improve the ability to predict the individuals with a high risk for CVD among asymptomatic type 2 diabetic patients. These results were compatible with a working hypothesis that assessment of tissue characteristics of carotid plaque as well as its size is useful for screening groups with a high risk of future cardiovascular event. To the best of our knowledge, this is the first report showing that the addition of GSM to conventional risk model improves the risk estimation for CVD event. Interestingly, in 184 (64%) out of 287 cases in this study, the thickest plaque also showed the lowest GSM value. Thus, when there were multiple carotid plaques in one individual, the thickest plaque was not always the plaque with the lowest GSM value. Therefore, it would be desirable to evaluate both the thickness and the echogenicity of each carotid plaque when assessing the risk for CVD. Several limitations of our study should be discussed. First, the number of study patients was relatively small, and there was a gender imbalance among participants. To conclusively demonstrate that the presence of echolucent carotid plaque is consistently and independently associated with cardiovascular events even after adjustment for conventional risk factors, many multivariate regression models should be tested. Unfortunately, such analyses were not able to be performed, since the number of patients who developed cardiovascular events during the observation period was not large enough. To compensate for this shortcoming, we adopted FRS as a risk prediction estimate that reflects accumulative effects of conventional risk factors. Although ROC curve analysis showed that the AUC of FRS for predicting risk of CVD in type 2 diabetic patients was only 0.60 in this study, this result was consistent with those of several previous studies, which also showed a value of around 0.60 [35,36]. Regardless, the clinical utility of ultrasonic tissue characterization of carotid plaque as a stratification tool for CVD risk should be verified in studies with larger sample size. Second, the precise cut-off level of GSM to detect individuals with high risk for CVD should be confirmed by further studies. We adopted the GSM values 0e5 for blood and 185e195 for adventitia, respectively, and found that the presence of GSM 37 plaque in the carotid was associated with higher incidence of CVD and that this cut-off level of GSM significantly improved the prediction ability for CVD. In similar settings, Falkowski et al. reported that plaques with GSM < 35 were associated with higher incidence of cerebral infarction [33], and Nicolaides et al. reported that cerebrovascular or retinal ischemic event occurred exclusively in patients with GSM < 30 [34]. Although these findings, including ours, indicated that GSM values in the range of 30e40 seem to be adequate cut-off values to detect individuals with high risk for CVD, this should be confirmed in further studies. Third, this study was limited to patients with carotid plaque. In the present study, 34 patients were not included in the analyses, because they had no carotid plaque. Interestingly, there were no incidences of CVD event in these 34 patients during the follow-up period, which suggests that patients without carotid plaque could be at low risk for CVD. Whether GSM values in carotid wall can provide any useful information for identifying individuals at high risk for CVD, even in patients without any carotid plaque, should be evaluated in another study.
Fourth, based on a cross-sectional analysis in 72 carotid endarterectomy specimens, Ciccone et al. reported that the degree of carotid plaque calcification was indicative of a high risk of coronary events [15], suggesting that the presence of calcified carotid plaques also serves as a marker of increased cardiovascular risk. Fifth, the ultrasound settings for each image were not always standardized. However, the blood was used as the reference for black and the adventitia as the reference for white, and gain settings for measurements within an individual were similar throughout the study. Therefore, the impact of gain of ultrasound beam on the GSM value would be quite small, if any. Sixth, the case of transient ischemic attack (n ¼ 1) should be excluded from the analyses, since it is not a hard event. However, a series of analyses showed similar results, regardless of whether this case was included. Seventh, the development of peripheral artery disease was considered as one of the components of primary outcome. However, in the present study, some asymptomatic peripheral artery disease cases might have been overlooked, since ankle-brachial indices were not measured in all the study patients. Finally, the present study focused on the question of whether measurement of GSM values can provide any useful information for identifying patients with a high risk for CVD, event at a certain point in their life with diabetes. Our thesis would have been strengthened if it were shown that measurement of GSM could improve the risk prediction of cardiovascular event after adjustment for the changes in cardio-metabolic control and therapeutic regimen during the follow-up period. However, we were not able to include many parameters into the multivariate regression model, because the number of patients in this study was not large enough. Whether longitudinal monitoring of these parameters during the follow-up period affects the prediction ability of GSM should be evaluated in further studies. Notwithstanding these limitations, our study indicates that measurement of carotid GSM can provide useful information for identifying patients with a high risk of CVD. In conclusion, the present study suggests that ultrasonic tissue characterization of carotid plaque using the GSM, a quantitative parameter of the plaque echogenicity, can improve the risk prediction of cardiovascular event in asymptomatic type 2 diabetic patients with carotid plaque. Author contributions Y.I., N.K., M.T. and H.K. researched data and wrote the manuscript. M.N., R.K., K.S., Y.Um., and S.S. researched the data. Y.Ue., K.K, and I.S. contributed to the interpretation of the results and the discussion. All authors reviewed and approved the report. N.K. is the guarantor of this work and, as such, had full access to all of the data in the study and takes responsibility for the integrity of the data and accuracy of the data analysis. Funding sources The Diabetes Masters Conference supported by the Japan Diabetes Foundation and Grants-in-Aid for Scientific Research from the Japanese Ministry of Education, Science, Sports, Culture and Technology [KAKENHI 25461349]. Conflict of interest disclosures None.
Y. Irie et al. / Atherosclerosis 230 (2013) 399e405
Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.atherosclerosis.2013.08.015. References [1] O’Leary DH, Polak JF, Kronmal RA, Manolio TA, Burke GL, Wolfson Jr SK. Carotid-artery intima and media thickness as a risk factor for myocardial infarction and stroke in older adults. Cardiovascular Health Study Collaborative Research Group. N Engl J Med 1999;340:14e22. [2] Nambi V, Chambless L, Folsom AR, et al. Carotid intima-media thickness and presence or absence of plaque improves prediction of coronary heart disease risk: the ARIC (Atherosclerosis Risk in Communities) study. J Am Coll Cardiol 2010;55:1600e7. [3] Polak JF, Pencina MJ, Pencina KM, O’Donnell CJ, Wolf PA, D’Agostino Sr RB. Carotid-wall intima-media thickness and cardiovascular events. N Engl J Med 2011;365:213e21. [4] Mathiesen EB, Johnsen SH, Wilsgaard T, Bønaa KH, Løchen ML, Njølstad I. Carotid plaque area and intima-media thickness in prediction of first-ever ischemic stroke: a 10-year follow-up of 6584 men and women: the Tromsø study. Stroke 2011;42:972e8. [5] Johnsen SH, Mathiesen EB, Joakimsen O, et al. Carotid atherosclerosis is a stronger predictor of myocardial infarction in women than in men: a 6-year follow-up study of 6226 persons: the Tromsø study. Stroke 2007;38:2873e80. [6] Fuster V, Badimon L, Badimon JJ, Chesebro JH. The pathogenesis of coronary artery disease and the acute coronary syndromes. N Engl J Med 1992;326:242e50. [7] Lee RT, Grodzinsky AJ, Frank EH, Kamm RD, Schoen FJ. Structure-dependent dynamic mechanical behavior of fibrous caps from human atherosclerotic plaques. Circulation 1991;83:1764e70. [8] Davies MJ, Richardson PD, Woolf N, Katz DR, Mann J. Risk of thrombosis in human atherosclerotic plaques: role of extracellular lipid, macrophage, and smooth muscle cell content. Br Heart J 1993;69:377e81. [9] Falk E. Pathogenesis of atherosclerosis. J Am Coll Cardiol 2006;47:C7e12. [10] Nishio M, Ueda Y, Matsuo K, et al. Detection of disrupted plaques by coronary CT: comparison with angioscopy. Heart 2011;97:1397e402. [11] Tahara N, Kai H, Nakaura H, et al. The prevalence of inflammation in carotid atherosclerosis: analysis with fluorodeoxyglucose-positron emission tomography. Eur Heart J 2007;28:2243e8. [12] Hatsukami TS, Ross R, Polissar NL, Yuan C. Visualization of fibrous cap thickness and rupture in human atherosclerotic carotid plaque in vivo with high-resolution magnetic resonance imaging. Circulation 2000;102:959e64. [13] General findings of the international atherosclerosis project. Lab Invest 1968;18:498e502. [14] Dalager S, Paaske WP, Kristensen IB, Laurberg JM, Falk E. Artery-related differences in atherosclerosis expression: implications for atherogenesis and dynamics in intima-media thickness. Stroke 2007;38:2698e705. [15] Ciccone MM, Marzullo A, Mizio D, et al. Can carotid plaque histology selectively predict the risk of an acute coronary syndrome? Int Heart J 2011;52:72e7. [16] Katakami N, Takahara M, Kaneto H, et al. Ultrasonic tissue characterization of carotid plaque improves the prediction of cardiovascular events in diabetic patients: a pilot study. Diabetes Care 2012;35:2640e6. [17] Grønholdt ML, Wiebe BM, Laursen H, Nielsen TG, Schroeder TV, Sillesen H. Lipid-rich carotid artery plaques appear echolucent on ultrasound B-mode images and may be associated with intraplaque haemorrhage. Eur J Vasc Endovasc Surg 1997;14:439e45. [18] Japan Diabetes Society. Diabetes mellitus treatment guideline based on scientific ground. Revised 2nd ed. Nankodo: Japan Diabetes Society; 2007. p. 257e72.
405
[19] D’Agostino Sr RB, Vasan RS, Pencina MJ, et al. General cardiovascular risk profile for use in primary care: the Framingham heart study. Circulation 2008;117:743e53. [20] Terminology and Diagnostic Criteria Committee, Japan Society of Ultrasonics in Medicine, Subcommittee for Preparing Guidelines for Ultrasound Diagnosis of Carotid Artery. Standard method for ultrasound evaluation of carotid artery lesions. Jpn J Med Ultrasonics 2009;36:501e18. [21] Sabetai MM, Tegos TJ, Nicolaides AN, Dhanjil S, Pare GJ, Stevens JM. Reproducibility of computer-quantified carotid plaque echogenicity: can we overcome the subjectivity? Stroke 2000;31:2189e96. [22] Smith Jr SC, Dove JT, Jacobs AK, et al., American College of Cardiology; American Heart Association Task Force on Practice Guidelines. Committee to Revise the 1993 Guidelines for Percutaneous Transluminal Coronary Angioplasty. ACC/AHA guidelines of percutaneous coronary interventions (revision of the 1993 PTCA guidelines) e executive summary. A report of the American College of Cardiology/American Heart Association task force on practice guidelines (committee to revise the 1993 guidelines for percutaneous transluminal coronary angioplasty). J Am Coll Cardiol 2001;37:2215e39. [23] Heagerty PJ, Lumley T, Pepe MS. Time-dependent ROC curves for censored survival data and a diagnostic marker. Biometrics 2000;56:337e44. [24] Salonen JT, Salonen R. Ultrasonographically assessed carotid morphology and the risk of coronary heart disease. Arterioscler Thromb 1991;11:1245e9. [25] Bots ML, Hoes AW, Koudstaal PJ, Hofman A, Grobbee DE. Common carotid intima-media thickness and risk of stroke and myocardial infarction: the Rotterdam study. Circulation 1997;96:1432e7. [26] Chambless LE, Folsom AR, Clrgg LX. Carotid wall thickness is predictive of incident clinical stroke: the Atherosclerosis Risk in Communities (ARIC) study. Am J Epidemiol 2000;151:478e87. [27] del Sol AI, Moons KG, Hollander M, et al. Is carotid intima-media thickness useful in cardiovascular disease risk assessment? The Rotterdam study. Stroke 2001;32:1532e8. [28] Lal BK, Hobson 2nd RW, Pappas PJ, et al. Pixel distribution analysis of B-mode ultrasound scan images predicts histologic features of atherosclerotic carotid plaques. J Vasc Surg 2002;35:1210e7. [29] Mathiesen EB, Bønaa KH, Joakimsen O. Echolucent plaques are associated with high risk of ischemic cerebrovascular events in carotid stenosis: the Tromsø study. Circulation 2001;103:2171e5. [30] Honda O, Sugiyama S, Kugiyama K, et al. Echolucent carotid plaques predict future coronary events in patients with coronary artery disease. J Am Coll Cardiol 2004;43:1177e84. [31] Seo Y, Watanabe S, Ishizu T, et al. Echolucent carotid plaques as a feature in patients with acute coronary syndrome. Circ J 2006;70:1629e34. [32] Grønholdt ML, Nordestgaard BG, Schroeder TV, Vorstrup S, Sillesen H. Ultrasonic echolucent carotid plaques predict future strokes. Circulation 2001;104: 68e73. [33] Falkowski A, Kaczmarczyk M, Cieszanowski A, Goracy I, Poncyliusz W, Wilk G. Computer-assisted characterisation of a carotid plaque. Med Sci Monit 2004;10:67e70. [34] Nicolaides AN, Kakkos SK, Kyriacou E, et al., Asymptomatic Carotid Stenosis and Risk of Stroke (ACSRS) Study Group. Asymptomatic internal carotid artery stenosis and cerebrovascular risk stratification. J Vasc Surg 2010;52:1486e96 e1e5. [35] van Dieren S, Peelen LM, Nöthlings U, et al. External validation of the UK prospective diabetes study (UKPDS) risk engine in patients with type 2 diabetes. Diabetologia 2011;54:264e70. [36] Kengne AP, Patel A, Colagiuri S, et al., ADVANCE Collaborative Group. The Framingham and UK prospective diabetes study (UKPDS) risk equations do not reliably estimate the probability of cardiovascular events in a large ethnically diverse sample of patients with diabetes: the action in diabetes and vascular disease: preterax and diamicron-MR controlled evaluation (ADVANCE) study. Diabetologia 2010;53:821e31.