- Email: [email protected]
Aortic Atherosclerosis Detected with Electron-Beam CT as a Predictor of Obstructive Coronary Artery Disease
Aortic Atherosclerosis Detected with ElectronBeam CT as a Predictor of Obstructive Coronary Artery Disease1 Junichiro Takasu, MD, PhD, Songshou Mao, MD, Matthew J. Budoff, MD
Rationale and Objectives. Several studies have demonstrated an association between coronary and aortic atherosclerosis. Aortic atherosclerosis is easily quantified by means of electron-beam computed tomography (CT). The aim of this study was to evaluate the usefulness of measurement of aortic atherosclerosis with electron-beam CT as an independent predictor of obstructive coronary artery disease (CAD). Materials and Methods. Ninety-seven patients (67 men, 30 women; mean age, 61 years ⫾ 12) were enrolled and underwent electron-beam CT with and without contrast material. Coronary artery calcification was quantified with nonenhanced electron-beam CT by means of Agatston score. CAD was defined as luminal narrowing of the coronary artery by at least 70%, as measured with electron-beam angiography. Aortic atherosclerosis was quantified by measuring raised lesions of the aortic wall (plaque) and wall thickening (volume and thickness) in the midportion of the descending thoracic aorta (10 contiguous sections), as depicted at contrast material– enhanced CT angiography. Results. Aortic plaque and calcification were detected only in patients who were at least 58 years old. The presence of aortic plaque was predictive of obstructive CAD, independent of coronary artery calcification. The sensitivity of aortic plaque (raised lesions) for obstructive CAD was 89% in patients at least 58 years old, and the specificity was 63%. Aortic calcification had a sensitivity of 56% and a specificity of 72% for diagnosis of obstructive CAD. Conclusion. This study demonstrated that aortic plaque detected with contrast-enhanced electron-beam CT was a more consistent predictor of obstructive CAD than other independent aortic variables. Aortic calcification depicted on nonenhanced CT images was highly specific for obstructive CAD. Key Words. Aorta, arteriosclerosis; aorta, CT; computed tomography (CT), electron beam; coronary vessels, calcification; coronary vessels, diseases. ©
AUR, 2003
Coronary artery calcification detected and measured by means of electron-beam computed tomography (CT) is a highly sensitive indicator of coronary artery disease (CAD) (1– 6). CT with contrast medium enhancement can
Acad Radiol 2003; 10:631– 637 1 From the Division of Cardiology, Harbor–UCLA Medical Center Research and Education Institute, 1124 W Carson St, Bldg RB-2, Torrance, CA 90502-2064. Received January 6, 2003; revision requested February 18; revision received March 3; accepted March 5. Address correspondence to M.J.B.
©
AUR, 2003
depict stenoses in large and medium-sized arteries (7–11). Aortic atherosclerosis is also easily visualized with contrast material– enhanced CT (7,8,10) and transesophageal echocardiography (12,13). Qualitative analysis categorizes CAD according to the presence of four types of atherosclerotic involvement: plaque, calcium deposits, wall thickening, and stenosis. Quantitative analysis includes measurement of wall thickness, wall volume, surface area of wall involvement, and luminal narrowing. Electronbeam CT can provide both qualitative and quantitative indicators of aortic wall involvement. Several studies have shown that the incidence of aortic atherosclerosis is
631
TAKASU ET AL
statistically significantly higher in patients with CAD (7,8,12,13). The aim of our study was to determine whether the aortic plaque burden measured by means of electron-beam CT with intravenous contrast medium enhancement is a specific marker for obstructive CAD. MATERIALS AND METHODS Patient Population Ninety-seven patients (67 men, 30 women; mean age, 61 years ⫾ 12) scheduled to undergo contrast-enhanced electron-beam CT for suspected CAD gave consent and were enrolled in this study from February 1998 to November 2001. Electron-beam CT angiography, which has been proved useful for detecting coronary stenosis (14 – 18), was performed in all patients. Patients were divided into two subgroups according to coronary stenosis grade at electron-beam CT angiography. Sixty had nonsignificant stenosis (luminal narrowing of ⬍70%), and 37 had a diagnosis of obstructive CAD (stenosis of ⱖ70%). Electron-Beam CT Methods The studies were performed with a C-150XP electronbeam CT scanner (GE-Imatron, South San Francisco, Calif) in high-resolution volume mode and with a 100msec exposure time, as previously described (2,19). Electrocardiographic triggering was used so that each image was obtained at the same point in diastole (20). Nonenhanced and Enhanced Aortic Images For nonenhanced scans, we used a standard protocol for cardiac CT imaging, with single-section acquisition and a high-resolution volume mode setting. The thoracic descending aorta was visualized without contrast material injection, and at least 30 consecutive images were obtained at 3-mm intervals and 3-mm table increments. Images were reconstructed with a 512 ⫻ 512 matrix and a 30-cm field of view, yielding a voxel size of 0.58 ⫻ 0.58 ⫻ 3 mm. The procedure for electron-beam CT coronary angiography has been previously described (14,15,18,21,22). Contrast-enhanced coronary volume scans were obtained with 3-mm collimation, 3-mm section thickness, and 2-mm table feed (ie, with no gap between sections) to provide overlap for optimal resolution of the coronary tree and aorta, and at least 55 consecutive images were obtained. Enhanced CT images had a 512 ⫻ 512 matrix and an 18-cm field of view, yielding a voxel size of 0.34 ⫻ 0.34 ⫻ 2 mm. The reconstruction kernel and al-
632
Academic Radiology, Vol 10, No 6, June 2003
gorithm were sharp and normal, respectively. Iopamidol, a nonionic contrast material (Bracco Diagnostics, Plainsboro, NJ), was administered through an antecubital vein with an injection rate of 4 mL/sec and a total volume of 120 –160 mL. Characteristics of Aortic Wall Involvement A single experienced investigator, blinded as to the patient’s clinical status, the electron-beam CT angiographic report, and the temporal relation of the scans, examined nonenhanced CT images for evidence of aortic atherosclerosis by using a commercially available workstation (AccuView; AccuImage Diagnostics, South San Francisco, Calif). Aortic plaque was defined as a raised lesion of the aortic wall or a thrombus, atheroma, or calcium deposit protruding into the lumen. Plaque was graded on a scale of 1– 4, according to the classification system devised by Fazio et al (12) for findings at transesophageal echocardiography and applied previously by Takasu et al (7) to findings at angiography. Aortic wall involvement in the midportion of the descending thoracic aorta was categorized as follows: (a) grade 1, no lumen irregularity and no wall thickening; (b) grade 2, wall thickening of more than 2 mm, without lumen irregularity; (c) grade 3, wall thickening with lumen irregularity, including plaques containing calcification of less than 0.10 cm3; and (d) grade 4, wall thickening with a high plaque burden and abundant calcification equal to or greater than 0.10 cm3 in volume (Fig 1). Aortic wall thickening was quantified by summing the wall volume measurements (in cubic centimeters) in the midportion of the descending aorta on 10 enhanced images. Aortic wall calcification (also measured in cubic centimeters) was defined as the volume of calcium deposits in the aortic midportion depicted on 10 nonenhanced images corresponding to the same sections as the enhanced images. Aortic wall thickness was defined as the maximum wall thickness (measured in millimeters) in the middle of the descending thoracic aorta. For each section, we investigated the thickness of the inner wall at the border of the aortic lumen and the thickness of the outer wall as depicted with CT. We used a commercially available workstation (AccuAnalyze; AccuImage Diagnostics) to calculate wall volume in cubic centimeters for each CT section. Statistical Analysis Continuous variables were quantified as means ⫾ standard deviations. The two-tailed t test was used to test
Academic Radiology, Vol 10, No 6, June 2003
CT FOR OBSTRUCTIVE CORONARY ARTERY DISEASE
Figure 1. Images of the aortic lumen, obtained with contrast-enhanced electron-beam CT angiography, show aortic wall involvement of (a) grade 1, (b) grade 2, (c) grade 3, and (d) grade 4 (12).
continuous variables in the coronary artery and aorta for correlation with CAD. Either the Fisher exact test or 2 analysis was used to compare categorical variables. Multivariate stepwise linear regression analysis was performed by using coronary artery calcification as the dependent variable. Logistic regression was used to determine the association between CAD and patient age, sex, coronary artery calcification, and aortic variables. Multivariate logistic regression analysis was performed with the forward stepwise likelihood ratio method to determine the independent variable for CAD. Sensitivity, specificity, and
predictive values were calculated with standard formulas. Accuracy was defined as the number of true-positive results plus the number of true-negative results divided by the total number of cases. Scores for coronary artery calcification and for aortic variables were transformed by using the natural log of (1 ⫹ score) in a secondary analysis. All statistical tests were two tailed, and P values less than .05 were considered to indicate statistically significant differences. Data were analyzed with SPSS software (version 8.0; SPSS, Chicago, Ill) in a Windows 98 environment (Microsoft, Redmond, Wash).
633
TAKASU ET AL
Academic Radiology, Vol 10, No 6, June 2003
Table 1 Patient Data as a Function of CAD
Figure 2. Patients with CAD (gray bars) or aortic plaque (black bars) as a function of age. Twenty patients were 49 years of age or younger, 23 were 50 –59 years, 26 were 60 – 69 years, and 28 were 70 years or older.
RESULTS Ninety-seven patients were enrolled (67 men, 30 women; mean age, 61 years ⫾ 12; range, 33– 82 years). Patients underwent both nonenhanced and enhanced electron-beam CT. Thirty-six patients (37%) had aortic plaque (grade 3 or 4), and 24 (25%) had aortic calcification. Figure 2 shows that patients with obstructive CAD were almost equally divided among the various age groups, whereas the prevalence of aortic plaque increased with age. The combination of aortic plaque and aortic calcification was detected only in patients aged 58 years or older. Table 1 shows patient data as a function of CAD. All variables were examined for the following three groups: all patients (n ⫽ 97), patients aged 58 years or older (“elder group,” n ⫽ 59), and patients aged less than 58 years (“younger group,” n ⫽ 38). The prevalence of coronary artery calcification was significantly increased in patients with obstructive CAD in all three groups (P ⬍ .001). In the elder group, the prevalences of aortic plaque, of wall thickening, and of aortic calcification were significantly higher in patients with obstructive CAD. All aortic wall variables were significantly associated with coronary artery calcification and with increasing age (P ⬍ .001). According to the results of multivariate stepwise linear regression analysis performed to determine which aortic variables were independently predictive of coronary artery calcification (Table 2), aortic plaque was an important independent variable both overall and in the elder group. In the younger group, aortic wall volume was an independent variable.
634
Variable
No CAD
CAD
P
No. of patients Age (y) Male (%) Coronary artery calcification Agatston score Volumetric score Aortic wall volume (cm3) Aortic calcification volume (cm3) Maximum aortic wall thickness (mm) Aortic plaque (%) Aortic calcification (%)
60 59 ⫾ 11 62
37 64 ⫾ 12 81
3.65 ⫾ 2.48 3.67 ⫾ 2.42 3.77 ⫾ 1.03
6.23 ⫾ 1.92 6.12 ⫾ 1.85 4.43 ⫾ 1.18
⬍.001 ⬍.001 .005
0.05 ⫾ 0.18
0.15 ⫾ 0.27
.061
2.7 ⫾ 1.1 20 15
3.5 ⫾ 1.6 65 41
.030 .040
.013 ⬍.001 .005
Note.—Means are given ⫾ standard deviations.
Patients were categorized according to coronary artery calcification score calculated with the Agatston method (Table 3). The first category included patients free of coronary artery calcification (score, 0), and only one of these patients had aortic plaque. For Agatston scores above 100, the sensitivity for aortic plaque was 81%, the specificity was 54%, the positive predictive value was 51%, the negative predictive value was 83%, and the accuracy was 64%. Table 4 shows the results of logistic regression analysis for the detection of obstructive CAD. The presence of aortic plaque in the elder group was associated with a risk of obstructive CAD 13 times higher than that for patients without aortic plaque. Multivariate logistic regression analysis demonstrated that coronary artery calcification (relative risk: 1.6; 95% confidence interval: 1.2, 2.1) and aortic plaque (relative risk: 4.0; 95% confidence interval: 1.5, 11.1) were independent predictors of obstructive CAD. Coronary artery calcification was an independent variable for obstructive CAD in all age groups. Among aortic variables, aortic plaque was an independent predictor of obstructive CAD for all patients and in the elder group. The sensitivities of aortic plaque and of aortic calcification, as seen with enhanced electron-beam CT, for the presence of obstructive CAD were 65% and 41%, respectively, for all patients and 89% and 56% for the elder group (Table 5). The specificities were 80% and 85% overall and 63% and 72% in the elder group. Table 6 shows the correlation between aortic calcification and aortic plaque. Table 7 shows the relation between aortic
CT FOR OBSTRUCTIVE CORONARY ARTERY DISEASE
Academic Radiology, Vol 10, No 6, June 2003
Table 2 Stepwise Linear Regression for Coronary Artery Calcification as Dependent Variable by Age, with Patient Sex and Aortic Plaque and Wall Volume as Independent Variables
Dependent Variable*
Independent Variable
Step
All patients Agatston score Volumetric score Patients aged ⱖ58 y Agatston score Volumetric score Patients aged ⬍58 y Agatston score Volumetric score
Standardized Coefficients 
P
Adjusted R2
1 2 1 2
Aortic plaque Sex Aortic plaque Sex
0.418 ⫺0.349 0.426 ⫺0.347
⬍.001 ⬍.001 ⬍.001 ⬍.001
0.166 0.282 0.161 0.275
1 2 1 2
Aortic plaque Sex Aortic plaque Sex
0.473 ⫺0.281 0.471 ⫺0.271
⬍.001 .014 ⬍.001 .019
0.215 0.282 0.212 0.274
1 2 1 2
Sex Aortic wall volume Sex Aortic wall volume
⫺0.373 0.308 ⫺0.370 0.335
.017 .047 .016 .028
0.205 0.271 0.212 0.295
*The Agatston and volumetric scores are scores for coronary artery calcification.
Table 3 Correlation between Aortic Plaque and Coronary Artery Calcification Agatston Score Aortic Plaque
⬍1
1–100
101–400
⬎400
Absent Present
15 1
18 6
12 5
16 24
Total
16
24
17
40
Note.—Data are numbers of patients.
plaque and obstructive CAD by age group. Accuracy was highest (85%) in patients aged 60 – 69 years. DISCUSSION In this study, we investigated the midportion of the descending thoracic aorta. The descending thoracic aorta has less atheromatous involvement than the infrarenal abdominal aorta but more than the ascending aorta (7,23). Many noninvasive studies of the thoracic aorta report a significant relationship between aortic atherosclerosis and CAD (7,12,13,24 –26). We investigated various characteristics of aortic involvement. Aortic wall thickness was a continuous characteristic of raised lesions of atherosclerosis, and aortic plaque and calcification were categorical
variables of raised lesions. We compared these two groups of aortic variables with coronary artery variables. Features of aortic raised lesions seen with electron-beam CT seem to be more closely associated with coronary artery variables than are surface features visible with other methods and modalities. Fazio et al (12) reported that patients with obstructive CAD had significantly more aortic plaque, as seen with transesophageal echocardiography (P ⬍ .001). In a pathologic study, Woolf et al (27) demonstrated that the presence of raised lesions of the aortic wall was more useful than the extent of intimal surface involvement for predicting obstructive CAD. These results suggest that detecting raised aortic lesions is more useful than quantifying aortic wall thickening. In our study, aortic plaque and calcification were seen only in patients aged 58 years or older. Obstructive CAD, however, was seen in patients as young as 37 years. Baker et al (28) demonstrated that severe coronary atherosclerosis was detected frequently at autopsy of individuals in their 40s and 50s and that aortic involvement rapidly increased after age 60 years. Our results correlate well with those of pathologic and transesophageal echocardiographic studies— especially studies performed in patients who were in their 6th decade or older. Patients who were free of coronary artery calcification were almost exclusively free of aortic plaque, but 29 (81%) of the 36 patients with aortic plaque had substantial coronary artery
635
TAKASU ET AL
Academic Radiology, Vol 10, No 6, June 2003
Table 4 Logistic Regression Analysis for CAD
Variable
Parameter Estimate
Standard Error
P
Relative Risk
95% Confidence Interval
Age Sex Agatston score* Volumetric score* Aortic wall volume Aortic calcification volume Maximum aortic wall thickness Aortic plaque Aortic calcification
0.042 ⫺0.980 0.579 0.595 0.553 1.983 0.430 1.999 1.352
0.020 0.497 0.142 0.150 0.206 1.042 0.166 0.472 0.493
.033 .049 ⬍.001 ⬍.001 .007 .057 .010 ⬍.001 .006
1.043 0.375 1.783 1.814 1.738 7.262 1.537 7.385 3.864
1.003–1.084 0.142–0.994 1.349–2.358 1.353–2.431 1.161–2.603 0.942–56.104 1.111–2.126 2.928–18.624 1.471–10.150
*The Agatston and volumetric scores are measures of coronary artery calcification.
Table 5 Correlation between Aortic Plaque, Aortic Calcification, and CAD
Table 6 Correlation between Aortic Plaque and Aortic Calcification Prediction of Aortic Plaque*
Prediction of Obstructive CAD* Sensitivity Specificity PPV NPV Accuracy Aortic Involvement (%) (%) (%) (%) (%) Aortic plaque All patients Patients aged ⱖ58 y Aortic calcification All patients Patients aged ⱖ58 y
65
80
67
79
74
89
63
67
87
75
41
85
63
70
68
56
72
63
66
64
Aortic Calcification All patients Patients aged ⱖ58 y
636
64
98
96
82
86
64
96
96
63
76
*PPV ⫽ positive predictive value, NPV ⫽ negative predictive value.
Table 7 Correlation between Aortic Plaque and CAD with Age
*PPV ⫽ positive predictive value, NPV ⫽ negative predictive value.
calcification (ie, an Agatston score ⬎ 100). These results verified a close relationship between atherosclerosis in the coronary artery and that in the aorta. Fazio et al (12) and Khoury et al (13) reported that aortic plaque had a sensitivity of 90% or more as a predictor of obstructive CAD. In our study, however, the sensitivity was relatively low (65%). These differences may be due to several factors. First, the prevalence of obstructive CAD was lower in our study (38%) than in these previous studies (63% and 67%, respectively). Second, we examined only a 25–30-mm portion of the descending thoracic aorta, excluding the ascending thoracic aorta, the arch, and the thoracoabdominal aorta. This may be why no aortic plaque was depicted in patients in the younger cohort. The sensitivity and specificity were 89% and 82%, respectively, in patients aged 60 – 69 years. The independent predictive value of aortic calcification in our
Sensitivity Specificity PPV NPV Accuracy (%) (%) (%) (%) (%)
Prediction of CAD* Age (y)
Sensitivity (%)
Specificity (%)
PPV (%)
NPV (%)
Accuracy (%)
33–49 50–59 60–69 70–82
0 29 89 93
100 100 82 31
... 100 73 61
70 76 93 80
70 78 85 64
*PPV ⫽ positive predictive value, NPV ⫽ negative predictive value.
study suggests that investigation of the aorta on nonenhanced cardiac studies is also useful for predicting obstructive CAD. Our study had several limitations. First, coronary stenosis was diagnosed by means of electron-beam CT angiography. The correlation between electron-beam CT angiography and conventional coronary angiography is high (14 –16,18) but not perfect. With electron-beam CT angiography, we realized a sensitivity of 96% for obstruc-
Academic Radiology, Vol 10, No 6, June 2003
tive CAD (18). The number of patients enrolled in our study was relatively small (n ⫽ 97). Moreover, we evaluated only the midthoracic aorta, and the descending thoracic aorta has less atheromatous involvement than the infrarenal abdominal aorta. Our results imply that aortic involvement detected with CT is an independent predictor of obstructive CAD in this symptomatic cohort. However, the generalizability of our findings will remain uncertain until larger studies are completed. In the Multiethnic Study of Atherosclerosis, this variable is being evaluated to see whether results are similar in an asymptomatic population. Aortic calcification is easily detected on nonenhanced images, has a 96% specificity for aortic plaque, and may be a marker for obstructive CAD. Our aortic data are most likely generalizable to multisection CT studies, in which the aorta is not subject to motion artifacts. Coronary artery calcification scores, however, have been shown to differ significantly between spiral CT and electron-beam CT, due to the slower image acquisition with spiral CT scanners (210 –300 msec) (29). The radiation dose to the patient from multisection CT is also up to sevenfold that from electron-beam CT (30,31). In conclusion, our study demonstrated that aortic plaque detected with enhanced electron-beam CT is a better independent predictor of obstructive CAD than are other variables of aortic thickening and stenosis. REFERENCES 1. Wexler L, Brundage B, Crouse J, et al. Coronary artery calcification: pathophysiology, epidemiology, imaging methods, and clinical implications—a statement for health professionals from the American Heart Association. Circulation 1996; 94:1175–1192. 2. Budoff MJ, Georgiou D, Brody A, et al. Ultrafast computed tomography as a diagnostic modality in the detection of coronary artery disease: a multicenter study. Circulation 1996; 93:898 –904. 3. Arad Y, Spadaro LA, Goodman K, et al. Prediction of coronary events with electron beam computed tomography. J Am Coll Cardiol 2000; 36:1253–1260. 4. Rumberger JA, Schwartz RS, Simons DB, et al. Relation of coronary calcium determined by electron beam computed tomography and lumen narrowing determined by autopsy. Am J Cardiol 1994; 73:1169 – 1173. 5. Schmermund A, Baumgart D, Gorge G, et al. Non-invasive visualization of coronary arteries with and without calcification by electron beam computed tomography. Herz 1996; 21:118 –126. 6. Nallamothu BK, Saint S, Bielak LF, et al. Electron-beam computed tomography in the diagnosis of coronary artery disease: a meta-analysis. Arch Intern Med 2001; 161:833– 838. 7. Takasu J, Takanashi K, Naito S, et al. Evaluation of morphological changes of the atherosclerotic aorta by enhanced computed tomography. Atherosclerosis 1992; 97:107–121. 8. Yamamoto R, Takasu J, Yokoyama K, et al. Descending aorta wall volume and coronary artery disease: a comparative study using enhanced computed tomography of the chest and coronary angiography. Jpn Circ J 2000; 64:842– 847.
CT FOR OBSTRUCTIVE CORONARY ARTERY DISEASE
9. Berg MH, Manninen HI, Rasanen HT, et al. CT angiography in the assessment of carotid artery atherosclerosis. Acta Radiol 2002; 43:116 – 124. 10. Ganaha F, Miller DC, Sugimoto K, et al. Prognosis of aortic intramural hematoma with and without penetrating atherosclerotic ulcer: a clinical and radiological analysis. Circulation 2002; 106:342–348. 11. Antiga L, Ene-Iordache B, Caverni L, et al. Geometric reconstruction for computational MeSH generation of arterial bifurcations from CT angiography. Comput Med Imaging Graph 2002; 26:227–235. 12. Fazio GP, Redberg RF, Winslow T, et al. Transesophageal echocardiographically detected atherosclerotic aortic plaque is a marker for coronary artery disease. J Am Coll Cardiol 1993; 21:144 –150. 13. Khoury Z, Schwartz R, Gottlieb S, et al. Relation of coronary artery disease to atherosclerotic disease in the aorta, carotid, and femoral arteries evaluated by ultrasound. Am J Cardiol 1997; 80:1429 –1433. 14. Achenbach S, Moshage W, Ropers D, et al. Value of electron-beam computed tomography for the noninvasive detection of high-grade coronary-artery stenoses and occlusions. N Engl J Med 1998; 339: 1964 –1971. 15. Budoff MJ, Oudiz RJ, Zalace CP, et al. Intravenous three-dimensional coronary angiography using contrast enhanced electron beam computed tomography. Am J Cardiol 1999; 83:840 – 845. 16. Achenbach S, Ropers D, Regenfus M, et al. Contrast enhanced electron beam computed tomography to analyse the coronary arteries in patients after acute myocardial infarction. Heart 2000; 84:489 – 493. 17. Ambrose JA, Tannenbaum MA, Alexopoulos D, et al. Angiographic progression of coronary artery disease and the development of myocardial infarction. J Am Coll Cardiol 1988; 12:56 – 62. 18. Lu B, Dai R, Bai H, et al. The effects of scanning and reconstruction parameters on image quality of three-dimensional electron-beam tomographic angiography: coronary artery phantom study. Acad Radiol 2000; 7:927–933. 19. Bakhsheshi H, Mao SS, Budoff MJ, et al. Preview method for electron beam scanning of the coronary arteries. Acad Radiol 2000; 7:620 – 626. 20. Mao SS, Budoff MJ, Bin L, Liu SC. Optimal ECG trigger point in electron-beam CT studies: three methods for minimizing motion artifacts. Acad Radiol 2001; 8:1107–1115. 21. Reddy GP, Chernoff DM, Adams JR, et al. Coronary artery stenoses: assessment with contrast-enhanced electron-beam CT and axial reconstructions. Radiology 1998; 208:167–172. 22. Schmermund A, Rensing BJ, Sheedy PF, et al. Intravenous electronbeam computed tomographic coronary angiography for segmental analysis of coronary artery stenoses. J Am Coll Cardiol 1998; 31:1547–1554. 23. Bjurulf P. Atherosclerosis and body-build with special reference to size and number of subcutaneous fat cells. Acta Med Scand 1959; 42:1– 99. 24. Tribouilloy C, Shen WF, Peltier M, et al. Noninvasive prediction of coronary artery disease by transesophageal echocardiographic detection of thoracic aortic plaque in valvular heart disease. Am J Cardiol 1994; 74:258 –260. 25. Belhassen L, Carville C, Pelle G, et al. Evaluation of carotid artery and aortic intima-media thickness measurements for exclusion of significant coronary atherosclerosis in patients scheduled for heart valve surgery. J Am Coll Cardiol 2002; 39:1139 –1144. 26. Agmon Y, Khandheria BK, Meissner I, et al. Relation of coronary artery disease and cerebrovascular disease with atherosclerosis of the thoracic aorta in the general population. Am J Cardiol 2002; 89:262–267. 27. Woolf N, Sacks MI, Davies MJ. Aortic plaque morphology in relation to coronary artery disease. Am J Pathol 1969; 57:187–197. 28. Baker SP, Spitz WU. Age effects and autopsy evidence of disease in fatally injured drivers. JAMA 1970; 214:1079 –1088. 29. Goldin JG, Yoon HC, Greaser LE III, et al. Spiral versus electron-beam CT for coronary artery calcium scoring. Radiology 2001; 221:213–221. 30. Morin RL, Gerber TC, McCollough CH. Radiation dose in computed tomography of the heart. Circulation 2003; 107:917–922. 31. Hunold P, Vogt FM, Schmermund A, et al. Radiation exposure during noninvasive coronary artery imaging: comparison of multislice CT and electron-beam CT (abstr). Radiology 2001; 221(P):540.
637
Rationale and Objectives. Several studies have demonstrated an association between coronary and aortic atherosclerosis. Aortic atherosclerosis is easily quantified by means of electron-beam computed tomography (CT). The aim of this study was to evaluate the usefulness of measurement of aortic atherosclerosis with electron-beam CT as an independent predictor of obstructive coronary artery disease (CAD). Materials and Methods. Ninety-seven patients (67 men, 30 women; mean age, 61 years ⫾ 12) were enrolled and underwent electron-beam CT with and without contrast material. Coronary artery calcification was quantified with nonenhanced electron-beam CT by means of Agatston score. CAD was defined as luminal narrowing of the coronary artery by at least 70%, as measured with electron-beam angiography. Aortic atherosclerosis was quantified by measuring raised lesions of the aortic wall (plaque) and wall thickening (volume and thickness) in the midportion of the descending thoracic aorta (10 contiguous sections), as depicted at contrast material– enhanced CT angiography. Results. Aortic plaque and calcification were detected only in patients who were at least 58 years old. The presence of aortic plaque was predictive of obstructive CAD, independent of coronary artery calcification. The sensitivity of aortic plaque (raised lesions) for obstructive CAD was 89% in patients at least 58 years old, and the specificity was 63%. Aortic calcification had a sensitivity of 56% and a specificity of 72% for diagnosis of obstructive CAD. Conclusion. This study demonstrated that aortic plaque detected with contrast-enhanced electron-beam CT was a more consistent predictor of obstructive CAD than other independent aortic variables. Aortic calcification depicted on nonenhanced CT images was highly specific for obstructive CAD. Key Words. Aorta, arteriosclerosis; aorta, CT; computed tomography (CT), electron beam; coronary vessels, calcification; coronary vessels, diseases. ©
AUR, 2003
Coronary artery calcification detected and measured by means of electron-beam computed tomography (CT) is a highly sensitive indicator of coronary artery disease (CAD) (1– 6). CT with contrast medium enhancement can
Acad Radiol 2003; 10:631– 637 1 From the Division of Cardiology, Harbor–UCLA Medical Center Research and Education Institute, 1124 W Carson St, Bldg RB-2, Torrance, CA 90502-2064. Received January 6, 2003; revision requested February 18; revision received March 3; accepted March 5. Address correspondence to M.J.B.
©
AUR, 2003
depict stenoses in large and medium-sized arteries (7–11). Aortic atherosclerosis is also easily visualized with contrast material– enhanced CT (7,8,10) and transesophageal echocardiography (12,13). Qualitative analysis categorizes CAD according to the presence of four types of atherosclerotic involvement: plaque, calcium deposits, wall thickening, and stenosis. Quantitative analysis includes measurement of wall thickness, wall volume, surface area of wall involvement, and luminal narrowing. Electronbeam CT can provide both qualitative and quantitative indicators of aortic wall involvement. Several studies have shown that the incidence of aortic atherosclerosis is
631
TAKASU ET AL
statistically significantly higher in patients with CAD (7,8,12,13). The aim of our study was to determine whether the aortic plaque burden measured by means of electron-beam CT with intravenous contrast medium enhancement is a specific marker for obstructive CAD. MATERIALS AND METHODS Patient Population Ninety-seven patients (67 men, 30 women; mean age, 61 years ⫾ 12) scheduled to undergo contrast-enhanced electron-beam CT for suspected CAD gave consent and were enrolled in this study from February 1998 to November 2001. Electron-beam CT angiography, which has been proved useful for detecting coronary stenosis (14 – 18), was performed in all patients. Patients were divided into two subgroups according to coronary stenosis grade at electron-beam CT angiography. Sixty had nonsignificant stenosis (luminal narrowing of ⬍70%), and 37 had a diagnosis of obstructive CAD (stenosis of ⱖ70%). Electron-Beam CT Methods The studies were performed with a C-150XP electronbeam CT scanner (GE-Imatron, South San Francisco, Calif) in high-resolution volume mode and with a 100msec exposure time, as previously described (2,19). Electrocardiographic triggering was used so that each image was obtained at the same point in diastole (20). Nonenhanced and Enhanced Aortic Images For nonenhanced scans, we used a standard protocol for cardiac CT imaging, with single-section acquisition and a high-resolution volume mode setting. The thoracic descending aorta was visualized without contrast material injection, and at least 30 consecutive images were obtained at 3-mm intervals and 3-mm table increments. Images were reconstructed with a 512 ⫻ 512 matrix and a 30-cm field of view, yielding a voxel size of 0.58 ⫻ 0.58 ⫻ 3 mm. The procedure for electron-beam CT coronary angiography has been previously described (14,15,18,21,22). Contrast-enhanced coronary volume scans were obtained with 3-mm collimation, 3-mm section thickness, and 2-mm table feed (ie, with no gap between sections) to provide overlap for optimal resolution of the coronary tree and aorta, and at least 55 consecutive images were obtained. Enhanced CT images had a 512 ⫻ 512 matrix and an 18-cm field of view, yielding a voxel size of 0.34 ⫻ 0.34 ⫻ 2 mm. The reconstruction kernel and al-
632
Academic Radiology, Vol 10, No 6, June 2003
gorithm were sharp and normal, respectively. Iopamidol, a nonionic contrast material (Bracco Diagnostics, Plainsboro, NJ), was administered through an antecubital vein with an injection rate of 4 mL/sec and a total volume of 120 –160 mL. Characteristics of Aortic Wall Involvement A single experienced investigator, blinded as to the patient’s clinical status, the electron-beam CT angiographic report, and the temporal relation of the scans, examined nonenhanced CT images for evidence of aortic atherosclerosis by using a commercially available workstation (AccuView; AccuImage Diagnostics, South San Francisco, Calif). Aortic plaque was defined as a raised lesion of the aortic wall or a thrombus, atheroma, or calcium deposit protruding into the lumen. Plaque was graded on a scale of 1– 4, according to the classification system devised by Fazio et al (12) for findings at transesophageal echocardiography and applied previously by Takasu et al (7) to findings at angiography. Aortic wall involvement in the midportion of the descending thoracic aorta was categorized as follows: (a) grade 1, no lumen irregularity and no wall thickening; (b) grade 2, wall thickening of more than 2 mm, without lumen irregularity; (c) grade 3, wall thickening with lumen irregularity, including plaques containing calcification of less than 0.10 cm3; and (d) grade 4, wall thickening with a high plaque burden and abundant calcification equal to or greater than 0.10 cm3 in volume (Fig 1). Aortic wall thickening was quantified by summing the wall volume measurements (in cubic centimeters) in the midportion of the descending aorta on 10 enhanced images. Aortic wall calcification (also measured in cubic centimeters) was defined as the volume of calcium deposits in the aortic midportion depicted on 10 nonenhanced images corresponding to the same sections as the enhanced images. Aortic wall thickness was defined as the maximum wall thickness (measured in millimeters) in the middle of the descending thoracic aorta. For each section, we investigated the thickness of the inner wall at the border of the aortic lumen and the thickness of the outer wall as depicted with CT. We used a commercially available workstation (AccuAnalyze; AccuImage Diagnostics) to calculate wall volume in cubic centimeters for each CT section. Statistical Analysis Continuous variables were quantified as means ⫾ standard deviations. The two-tailed t test was used to test
Academic Radiology, Vol 10, No 6, June 2003
CT FOR OBSTRUCTIVE CORONARY ARTERY DISEASE
Figure 1. Images of the aortic lumen, obtained with contrast-enhanced electron-beam CT angiography, show aortic wall involvement of (a) grade 1, (b) grade 2, (c) grade 3, and (d) grade 4 (12).
continuous variables in the coronary artery and aorta for correlation with CAD. Either the Fisher exact test or 2 analysis was used to compare categorical variables. Multivariate stepwise linear regression analysis was performed by using coronary artery calcification as the dependent variable. Logistic regression was used to determine the association between CAD and patient age, sex, coronary artery calcification, and aortic variables. Multivariate logistic regression analysis was performed with the forward stepwise likelihood ratio method to determine the independent variable for CAD. Sensitivity, specificity, and
predictive values were calculated with standard formulas. Accuracy was defined as the number of true-positive results plus the number of true-negative results divided by the total number of cases. Scores for coronary artery calcification and for aortic variables were transformed by using the natural log of (1 ⫹ score) in a secondary analysis. All statistical tests were two tailed, and P values less than .05 were considered to indicate statistically significant differences. Data were analyzed with SPSS software (version 8.0; SPSS, Chicago, Ill) in a Windows 98 environment (Microsoft, Redmond, Wash).
633
TAKASU ET AL
Academic Radiology, Vol 10, No 6, June 2003
Table 1 Patient Data as a Function of CAD
Figure 2. Patients with CAD (gray bars) or aortic plaque (black bars) as a function of age. Twenty patients were 49 years of age or younger, 23 were 50 –59 years, 26 were 60 – 69 years, and 28 were 70 years or older.
RESULTS Ninety-seven patients were enrolled (67 men, 30 women; mean age, 61 years ⫾ 12; range, 33– 82 years). Patients underwent both nonenhanced and enhanced electron-beam CT. Thirty-six patients (37%) had aortic plaque (grade 3 or 4), and 24 (25%) had aortic calcification. Figure 2 shows that patients with obstructive CAD were almost equally divided among the various age groups, whereas the prevalence of aortic plaque increased with age. The combination of aortic plaque and aortic calcification was detected only in patients aged 58 years or older. Table 1 shows patient data as a function of CAD. All variables were examined for the following three groups: all patients (n ⫽ 97), patients aged 58 years or older (“elder group,” n ⫽ 59), and patients aged less than 58 years (“younger group,” n ⫽ 38). The prevalence of coronary artery calcification was significantly increased in patients with obstructive CAD in all three groups (P ⬍ .001). In the elder group, the prevalences of aortic plaque, of wall thickening, and of aortic calcification were significantly higher in patients with obstructive CAD. All aortic wall variables were significantly associated with coronary artery calcification and with increasing age (P ⬍ .001). According to the results of multivariate stepwise linear regression analysis performed to determine which aortic variables were independently predictive of coronary artery calcification (Table 2), aortic plaque was an important independent variable both overall and in the elder group. In the younger group, aortic wall volume was an independent variable.
634
Variable
No CAD
CAD
P
No. of patients Age (y) Male (%) Coronary artery calcification Agatston score Volumetric score Aortic wall volume (cm3) Aortic calcification volume (cm3) Maximum aortic wall thickness (mm) Aortic plaque (%) Aortic calcification (%)
60 59 ⫾ 11 62
37 64 ⫾ 12 81
3.65 ⫾ 2.48 3.67 ⫾ 2.42 3.77 ⫾ 1.03
6.23 ⫾ 1.92 6.12 ⫾ 1.85 4.43 ⫾ 1.18
⬍.001 ⬍.001 .005
0.05 ⫾ 0.18
0.15 ⫾ 0.27
.061
2.7 ⫾ 1.1 20 15
3.5 ⫾ 1.6 65 41
.030 .040
.013 ⬍.001 .005
Note.—Means are given ⫾ standard deviations.
Patients were categorized according to coronary artery calcification score calculated with the Agatston method (Table 3). The first category included patients free of coronary artery calcification (score, 0), and only one of these patients had aortic plaque. For Agatston scores above 100, the sensitivity for aortic plaque was 81%, the specificity was 54%, the positive predictive value was 51%, the negative predictive value was 83%, and the accuracy was 64%. Table 4 shows the results of logistic regression analysis for the detection of obstructive CAD. The presence of aortic plaque in the elder group was associated with a risk of obstructive CAD 13 times higher than that for patients without aortic plaque. Multivariate logistic regression analysis demonstrated that coronary artery calcification (relative risk: 1.6; 95% confidence interval: 1.2, 2.1) and aortic plaque (relative risk: 4.0; 95% confidence interval: 1.5, 11.1) were independent predictors of obstructive CAD. Coronary artery calcification was an independent variable for obstructive CAD in all age groups. Among aortic variables, aortic plaque was an independent predictor of obstructive CAD for all patients and in the elder group. The sensitivities of aortic plaque and of aortic calcification, as seen with enhanced electron-beam CT, for the presence of obstructive CAD were 65% and 41%, respectively, for all patients and 89% and 56% for the elder group (Table 5). The specificities were 80% and 85% overall and 63% and 72% in the elder group. Table 6 shows the correlation between aortic calcification and aortic plaque. Table 7 shows the relation between aortic
CT FOR OBSTRUCTIVE CORONARY ARTERY DISEASE
Academic Radiology, Vol 10, No 6, June 2003
Table 2 Stepwise Linear Regression for Coronary Artery Calcification as Dependent Variable by Age, with Patient Sex and Aortic Plaque and Wall Volume as Independent Variables
Dependent Variable*
Independent Variable
Step
All patients Agatston score Volumetric score Patients aged ⱖ58 y Agatston score Volumetric score Patients aged ⬍58 y Agatston score Volumetric score
Standardized Coefficients 
P
Adjusted R2
1 2 1 2
Aortic plaque Sex Aortic plaque Sex
0.418 ⫺0.349 0.426 ⫺0.347
⬍.001 ⬍.001 ⬍.001 ⬍.001
0.166 0.282 0.161 0.275
1 2 1 2
Aortic plaque Sex Aortic plaque Sex
0.473 ⫺0.281 0.471 ⫺0.271
⬍.001 .014 ⬍.001 .019
0.215 0.282 0.212 0.274
1 2 1 2
Sex Aortic wall volume Sex Aortic wall volume
⫺0.373 0.308 ⫺0.370 0.335
.017 .047 .016 .028
0.205 0.271 0.212 0.295
*The Agatston and volumetric scores are scores for coronary artery calcification.
Table 3 Correlation between Aortic Plaque and Coronary Artery Calcification Agatston Score Aortic Plaque
⬍1
1–100
101–400
⬎400
Absent Present
15 1
18 6
12 5
16 24
Total
16
24
17
40
Note.—Data are numbers of patients.
plaque and obstructive CAD by age group. Accuracy was highest (85%) in patients aged 60 – 69 years. DISCUSSION In this study, we investigated the midportion of the descending thoracic aorta. The descending thoracic aorta has less atheromatous involvement than the infrarenal abdominal aorta but more than the ascending aorta (7,23). Many noninvasive studies of the thoracic aorta report a significant relationship between aortic atherosclerosis and CAD (7,12,13,24 –26). We investigated various characteristics of aortic involvement. Aortic wall thickness was a continuous characteristic of raised lesions of atherosclerosis, and aortic plaque and calcification were categorical
variables of raised lesions. We compared these two groups of aortic variables with coronary artery variables. Features of aortic raised lesions seen with electron-beam CT seem to be more closely associated with coronary artery variables than are surface features visible with other methods and modalities. Fazio et al (12) reported that patients with obstructive CAD had significantly more aortic plaque, as seen with transesophageal echocardiography (P ⬍ .001). In a pathologic study, Woolf et al (27) demonstrated that the presence of raised lesions of the aortic wall was more useful than the extent of intimal surface involvement for predicting obstructive CAD. These results suggest that detecting raised aortic lesions is more useful than quantifying aortic wall thickening. In our study, aortic plaque and calcification were seen only in patients aged 58 years or older. Obstructive CAD, however, was seen in patients as young as 37 years. Baker et al (28) demonstrated that severe coronary atherosclerosis was detected frequently at autopsy of individuals in their 40s and 50s and that aortic involvement rapidly increased after age 60 years. Our results correlate well with those of pathologic and transesophageal echocardiographic studies— especially studies performed in patients who were in their 6th decade or older. Patients who were free of coronary artery calcification were almost exclusively free of aortic plaque, but 29 (81%) of the 36 patients with aortic plaque had substantial coronary artery
635
TAKASU ET AL
Academic Radiology, Vol 10, No 6, June 2003
Table 4 Logistic Regression Analysis for CAD
Variable
Parameter Estimate
Standard Error
P
Relative Risk
95% Confidence Interval
Age Sex Agatston score* Volumetric score* Aortic wall volume Aortic calcification volume Maximum aortic wall thickness Aortic plaque Aortic calcification
0.042 ⫺0.980 0.579 0.595 0.553 1.983 0.430 1.999 1.352
0.020 0.497 0.142 0.150 0.206 1.042 0.166 0.472 0.493
.033 .049 ⬍.001 ⬍.001 .007 .057 .010 ⬍.001 .006
1.043 0.375 1.783 1.814 1.738 7.262 1.537 7.385 3.864
1.003–1.084 0.142–0.994 1.349–2.358 1.353–2.431 1.161–2.603 0.942–56.104 1.111–2.126 2.928–18.624 1.471–10.150
*The Agatston and volumetric scores are measures of coronary artery calcification.
Table 5 Correlation between Aortic Plaque, Aortic Calcification, and CAD
Table 6 Correlation between Aortic Plaque and Aortic Calcification Prediction of Aortic Plaque*
Prediction of Obstructive CAD* Sensitivity Specificity PPV NPV Accuracy Aortic Involvement (%) (%) (%) (%) (%) Aortic plaque All patients Patients aged ⱖ58 y Aortic calcification All patients Patients aged ⱖ58 y
65
80
67
79
74
89
63
67
87
75
41
85
63
70
68
56
72
63
66
64
Aortic Calcification All patients Patients aged ⱖ58 y
636
64
98
96
82
86
64
96
96
63
76
*PPV ⫽ positive predictive value, NPV ⫽ negative predictive value.
Table 7 Correlation between Aortic Plaque and CAD with Age
*PPV ⫽ positive predictive value, NPV ⫽ negative predictive value.
calcification (ie, an Agatston score ⬎ 100). These results verified a close relationship between atherosclerosis in the coronary artery and that in the aorta. Fazio et al (12) and Khoury et al (13) reported that aortic plaque had a sensitivity of 90% or more as a predictor of obstructive CAD. In our study, however, the sensitivity was relatively low (65%). These differences may be due to several factors. First, the prevalence of obstructive CAD was lower in our study (38%) than in these previous studies (63% and 67%, respectively). Second, we examined only a 25–30-mm portion of the descending thoracic aorta, excluding the ascending thoracic aorta, the arch, and the thoracoabdominal aorta. This may be why no aortic plaque was depicted in patients in the younger cohort. The sensitivity and specificity were 89% and 82%, respectively, in patients aged 60 – 69 years. The independent predictive value of aortic calcification in our
Sensitivity Specificity PPV NPV Accuracy (%) (%) (%) (%) (%)
Prediction of CAD* Age (y)
Sensitivity (%)
Specificity (%)
PPV (%)
NPV (%)
Accuracy (%)
33–49 50–59 60–69 70–82
0 29 89 93
100 100 82 31
... 100 73 61
70 76 93 80
70 78 85 64
*PPV ⫽ positive predictive value, NPV ⫽ negative predictive value.
study suggests that investigation of the aorta on nonenhanced cardiac studies is also useful for predicting obstructive CAD. Our study had several limitations. First, coronary stenosis was diagnosed by means of electron-beam CT angiography. The correlation between electron-beam CT angiography and conventional coronary angiography is high (14 –16,18) but not perfect. With electron-beam CT angiography, we realized a sensitivity of 96% for obstruc-
Academic Radiology, Vol 10, No 6, June 2003
tive CAD (18). The number of patients enrolled in our study was relatively small (n ⫽ 97). Moreover, we evaluated only the midthoracic aorta, and the descending thoracic aorta has less atheromatous involvement than the infrarenal abdominal aorta. Our results imply that aortic involvement detected with CT is an independent predictor of obstructive CAD in this symptomatic cohort. However, the generalizability of our findings will remain uncertain until larger studies are completed. In the Multiethnic Study of Atherosclerosis, this variable is being evaluated to see whether results are similar in an asymptomatic population. Aortic calcification is easily detected on nonenhanced images, has a 96% specificity for aortic plaque, and may be a marker for obstructive CAD. Our aortic data are most likely generalizable to multisection CT studies, in which the aorta is not subject to motion artifacts. Coronary artery calcification scores, however, have been shown to differ significantly between spiral CT and electron-beam CT, due to the slower image acquisition with spiral CT scanners (210 –300 msec) (29). The radiation dose to the patient from multisection CT is also up to sevenfold that from electron-beam CT (30,31). In conclusion, our study demonstrated that aortic plaque detected with enhanced electron-beam CT is a better independent predictor of obstructive CAD than are other variables of aortic thickening and stenosis. REFERENCES 1. Wexler L, Brundage B, Crouse J, et al. Coronary artery calcification: pathophysiology, epidemiology, imaging methods, and clinical implications—a statement for health professionals from the American Heart Association. Circulation 1996; 94:1175–1192. 2. Budoff MJ, Georgiou D, Brody A, et al. Ultrafast computed tomography as a diagnostic modality in the detection of coronary artery disease: a multicenter study. Circulation 1996; 93:898 –904. 3. Arad Y, Spadaro LA, Goodman K, et al. Prediction of coronary events with electron beam computed tomography. J Am Coll Cardiol 2000; 36:1253–1260. 4. Rumberger JA, Schwartz RS, Simons DB, et al. Relation of coronary calcium determined by electron beam computed tomography and lumen narrowing determined by autopsy. Am J Cardiol 1994; 73:1169 – 1173. 5. Schmermund A, Baumgart D, Gorge G, et al. Non-invasive visualization of coronary arteries with and without calcification by electron beam computed tomography. Herz 1996; 21:118 –126. 6. Nallamothu BK, Saint S, Bielak LF, et al. Electron-beam computed tomography in the diagnosis of coronary artery disease: a meta-analysis. Arch Intern Med 2001; 161:833– 838. 7. Takasu J, Takanashi K, Naito S, et al. Evaluation of morphological changes of the atherosclerotic aorta by enhanced computed tomography. Atherosclerosis 1992; 97:107–121. 8. Yamamoto R, Takasu J, Yokoyama K, et al. Descending aorta wall volume and coronary artery disease: a comparative study using enhanced computed tomography of the chest and coronary angiography. Jpn Circ J 2000; 64:842– 847.
CT FOR OBSTRUCTIVE CORONARY ARTERY DISEASE
9. Berg MH, Manninen HI, Rasanen HT, et al. CT angiography in the assessment of carotid artery atherosclerosis. Acta Radiol 2002; 43:116 – 124. 10. Ganaha F, Miller DC, Sugimoto K, et al. Prognosis of aortic intramural hematoma with and without penetrating atherosclerotic ulcer: a clinical and radiological analysis. Circulation 2002; 106:342–348. 11. Antiga L, Ene-Iordache B, Caverni L, et al. Geometric reconstruction for computational MeSH generation of arterial bifurcations from CT angiography. Comput Med Imaging Graph 2002; 26:227–235. 12. Fazio GP, Redberg RF, Winslow T, et al. Transesophageal echocardiographically detected atherosclerotic aortic plaque is a marker for coronary artery disease. J Am Coll Cardiol 1993; 21:144 –150. 13. Khoury Z, Schwartz R, Gottlieb S, et al. Relation of coronary artery disease to atherosclerotic disease in the aorta, carotid, and femoral arteries evaluated by ultrasound. Am J Cardiol 1997; 80:1429 –1433. 14. Achenbach S, Moshage W, Ropers D, et al. Value of electron-beam computed tomography for the noninvasive detection of high-grade coronary-artery stenoses and occlusions. N Engl J Med 1998; 339: 1964 –1971. 15. Budoff MJ, Oudiz RJ, Zalace CP, et al. Intravenous three-dimensional coronary angiography using contrast enhanced electron beam computed tomography. Am J Cardiol 1999; 83:840 – 845. 16. Achenbach S, Ropers D, Regenfus M, et al. Contrast enhanced electron beam computed tomography to analyse the coronary arteries in patients after acute myocardial infarction. Heart 2000; 84:489 – 493. 17. Ambrose JA, Tannenbaum MA, Alexopoulos D, et al. Angiographic progression of coronary artery disease and the development of myocardial infarction. J Am Coll Cardiol 1988; 12:56 – 62. 18. Lu B, Dai R, Bai H, et al. The effects of scanning and reconstruction parameters on image quality of three-dimensional electron-beam tomographic angiography: coronary artery phantom study. Acad Radiol 2000; 7:927–933. 19. Bakhsheshi H, Mao SS, Budoff MJ, et al. Preview method for electron beam scanning of the coronary arteries. Acad Radiol 2000; 7:620 – 626. 20. Mao SS, Budoff MJ, Bin L, Liu SC. Optimal ECG trigger point in electron-beam CT studies: three methods for minimizing motion artifacts. Acad Radiol 2001; 8:1107–1115. 21. Reddy GP, Chernoff DM, Adams JR, et al. Coronary artery stenoses: assessment with contrast-enhanced electron-beam CT and axial reconstructions. Radiology 1998; 208:167–172. 22. Schmermund A, Rensing BJ, Sheedy PF, et al. Intravenous electronbeam computed tomographic coronary angiography for segmental analysis of coronary artery stenoses. J Am Coll Cardiol 1998; 31:1547–1554. 23. Bjurulf P. Atherosclerosis and body-build with special reference to size and number of subcutaneous fat cells. Acta Med Scand 1959; 42:1– 99. 24. Tribouilloy C, Shen WF, Peltier M, et al. Noninvasive prediction of coronary artery disease by transesophageal echocardiographic detection of thoracic aortic plaque in valvular heart disease. Am J Cardiol 1994; 74:258 –260. 25. Belhassen L, Carville C, Pelle G, et al. Evaluation of carotid artery and aortic intima-media thickness measurements for exclusion of significant coronary atherosclerosis in patients scheduled for heart valve surgery. J Am Coll Cardiol 2002; 39:1139 –1144. 26. Agmon Y, Khandheria BK, Meissner I, et al. Relation of coronary artery disease and cerebrovascular disease with atherosclerosis of the thoracic aorta in the general population. Am J Cardiol 2002; 89:262–267. 27. Woolf N, Sacks MI, Davies MJ. Aortic plaque morphology in relation to coronary artery disease. Am J Pathol 1969; 57:187–197. 28. Baker SP, Spitz WU. Age effects and autopsy evidence of disease in fatally injured drivers. JAMA 1970; 214:1079 –1088. 29. Goldin JG, Yoon HC, Greaser LE III, et al. Spiral versus electron-beam CT for coronary artery calcium scoring. Radiology 2001; 221:213–221. 30. Morin RL, Gerber TC, McCollough CH. Radiation dose in computed tomography of the heart. Circulation 2003; 107:917–922. 31. Hunold P, Vogt FM, Schmermund A, et al. Radiation exposure during noninvasive coronary artery imaging: comparison of multislice CT and electron-beam CT (abstr). Radiology 2001; 221(P):540.
637