Coronary computed tomographic (CT) angiography has been proposed as an alternative diagnostic modality for assessing obstructive coronary artery disea

Transcription

1 Note: This copy is for your personal, non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at ORIGINAL RESEARCH n CARDIAC IMAGING Coronary Artery Calcium Scoring Does Not Add Prognostic Value to Standard 64-Section CT Angiography Protocol in Low-Risk Patients Suspected of Having Coronary Artery Disease1 Sung Woo Kwon, MD Young Jin Kim, MD Jaemin Shim, MD Ji Min Sung, PhD Mi Eun Han, BS Dong Won Kang, BS Ji-Ye Kim Byoung Wook Choi, MD Hyuk-Jae Chang, MD 1 From the Divisions of Cardiology (S.W.K., J.S., M.E.H., D.W.K., J.Y.K., H.J.C.) and Radiology (Y.J.K., B.W.C.), Severance Cardiovascular Hospital, and Department of Research Affairs (J.M.S.), Yonsei University College of Medicine, 250 Seongsanno Seodaemungu, Seoul , Republic of Korea. Received May 5, 2010; revision requested June 10; revision received September 13; accepted September 28; fi nal version accepted November 2. Supported by a grant (grant A102064) from the Korean Health Technology R&D Project, Ministry for Health and Welfare, Seoul, Republic of Korea. Address correspondence to H.J.C. ( hjchang@yuhs.ac ). Purpose: Materials and Methods: Results: Conclusion: To evaluate the prognostic outcome of cardiac computed tomography (CT) for prediction of major adverse cardiac events (MACEs) in low-risk patients suspected of having coronary artery disease (CAD) and to explore the differential prognostic values of coronary artery calcium (CAC) scoring and coronary CT angiography. Institutional review committee approval and informed consent were obtained. In 4338 patients who underwent 64-section CT for evaluation of suspected CAD, both CAC scoring and CT angiography were concurrently performed by using standard scanning protocols. Follow-up clinical outcome data regarding composite MACEs were procured. Multivariable Cox proportional hazards models were developed to predict MACEs. Risk-adjusted models incorporated traditional risk factors for CAC scoring and coronary CT angiography. During the mean follow-up of 828 days 6 380, there were 105 MACEs, for an event rate of 3%. The presence of obstructive CAD at coronary CT angiography had independent prognostic value, which escalated according to the number of stenosed vessels ( P,.001). In the receiver operating characteristic curve (ROC) analysis, the superiority of coronary CT angiography to CAC scoring was demonstrated by a significantly greater area under the ROC curve (AUC) (0.892 vs 0.810, P,.001), whereas no significant incremental value for the addition of CAC scoring to coronary CT angiography was established (AUC = for coronary CT angiography alone vs with addition of CAC scoring, P =.198). Coronary CT angiography is better than CAC scoring in predicting MACEs in low-risk patients suspected of having CAD. Furthermore, the current standard multisection CT protocol (coronary CT angiography combined with CAC scoring) has no incremental prognostic value compared with coronary CT angiography alone. Therefore, in terms of determining prognosis, CAC scoring may no longer need to be incorporated in the cardiac CT protocol in this population. q RSNA, 2011 q RSNA, radiology.rsna.org n Radiology: Volume 259: Number 1 April 2011

2 Coronary computed tomographic (CT) angiography has been proposed as an alternative diagnostic modality for assessing obstructive coronary artery disease (CAD) in cohorts with diverse risk profiles for coronary heart disease ( 1 ). Up to the present, coronary artery calcium (CAC) scores have been accepted as an effective prognostic indicator of coronary heart disease, independent of traditional risk factors ( 2 5 ). Although CAC scores are a reliable surrogate marker of atherosclerosis as a representation of plaque burden, their correlation with results of invasive coronary angiography, the current reference standard for diagnosing obstructive CAD, is limited in terms of detecting significant stenosis ( 6 ). In contrast, coronary CT angiography has high diagnostic accuracy for depicting significant obstructive CAD ( 7 ). Currently, performing both CAC scoring and coronary CT angiography in combination is the conventional protocol for diagnostic evaluation of CAD ( 8 10 ). Recent data ( 11 ) have demonstrated the additive prognostic value of CAC scoring to coronary CT angiography performed by using electronbeam CT in patients suspected of having CAD. Nonetheless, to date, to our knowledge, no comprehensive study has been performed to decide the optimal protocol for current multidetector CT technology in terms of predicting Advances in Knowledge n Coronary CT angiography data correlate positively with coronary artery calcium (CAC) scoring data and significantly predict the risk of major adverse cardiac events. n Coronary CT angiography has a better predictive value than CAC scoring, and, moreover, the standard multisection CT protocol (coronary CT angiography combined with CAC scoring) confers no added prognostic value compared with coronary CT angiography alone in low-risk patients suspected of having coronary artery disease (CAD). outcomes in patients suspected of having CAD. The purpose of this study was to evaluate the prognostic outcome of cardiac CT for prediction of major adverse cardiac events (MACEs) in lowrisk patients suspected of having CAD and to explore the differential prognostic values of CAC scoring and coronary CT angiography. Materials and Methods Study Design and Patient Selection Institutional review committee approval and informed consent were obtained. This was an observational single-center study in which 4338 consecutive patients 30 years of age or older (2299 men [53%]; mean age, 60 years 6 10 [standard deviation]) with no prior documented CAD underwent 64-section CAC scoring and coronary CT angiography between May 2003 and April 2009 at Severance Hospital for the evaluation of suspected CAD. Imaging indications included evaluation of symptoms; evaluation of signs of cardiac disease such as abnormal resting or stress electrocardiographic (ECG) test results; and evaluation of asymptomatic patients with peripheral arterial disease, cerebrovascular disease, or multiple CAD risk factors. Symptoms included typical angina, atypical angina, nonanginal chest pain, and dyspnea. The pretest likelihood of CAD was determined on the basis of the American College of Cardiology/American Heart Association guidelines ( 12 ). Patients with stable conditions who underwent elective revascularization within 60 days after the index multisection CT examination were excluded from the analysis. Risk factors were recorded as Implication for Patient Care n In terms of prognosis, CAC scor- ing may no longer need to be incorporated in the cardiac CT protocol in low-risk patients suspected of having CAD because its disadvantages in terms of radiation exposure outweigh its value in terms of prognostication. categoric variables in all patients. Hypertension was considered to be present with a systolic blood pressure of 140 mm Hg or greater and/or a diastolic blood pressure of 90 mm Hg or greater or treatment with antihypertensive agents. Diabetes was considered to be present when the patient was receiving treatment with hypoglycemic agents or insulin, had a fasting glucose level of 126 mg/dl (7.0 mmol/l) or greater, or had known but untreated hyperglycemia. A history of smoking was considered to be present if patients currently smoked or if they smoked until 1 month before the study. Dyslipidemia was defined as a total cholesterol level of 240 mg/dl (6.21 mmol/l) or greater, a low-density lipoprotein level of 130 mg/dl (3.37 mmol/l) or greater, a high-density lipoprotein level of less than 40 mg/dl (1.0 mmol/l), a triglyceride level of 200 mg/dl (2.26 mmol/l) or greater, and/or treatment with lipid-lowering agents. Laboratory values for lipid parameters were obtained within 1 month prior to the CT examination. Published online before print /radiol Radiology 2011; 259:92 99 Abbreviations: AUC = area under the ROC curve CAC = coronary artery calcium CAD = coronary artery disease CI = confi dence interval ECG = electrocardiography HR = hazard ratio MACE = major adverse cardiac event MI = myocardial infarction ROC = receiver operating characteristic Author contributions: Guarantors of integrity of entire study, S.W.K., J.M.S., M.E.H., D.W.K., J.Y.K., H.J.C.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript fi nal version approval, all authors; literature research, S.W.K., Y.J.K., J.S., M.E.H., D.W.K., J.Y.K., H.J.C.; clinical studies, S.W.K., Y.J.K., J.S., J.M.S., M.E.H., D.W.K., B.W.C., H.J.C.; statistical analysis, S.W.K., Y.J.K., J.S., J.M.S., M.E.H., D.W.K., H.J.C.; and manuscript editing, S.W.K., Y.J.K., J.S., M.E.H., D.W.K., J.Y.K., B.W.C., H.J.C. Potential confl icts of interest are listed at the end of this article. Radiology: Volume 259: Number 1 April 2011 n radiology.rsna.org 93

3 CT Protocol Before the CT examination, the heart rate of the patient was monitored. Patients without a contraindication to b -adrenergic blocking agents and/or calcium channel blocking agents (eg, overt heart failure, atrioventricular conduction abnormalities) and with initial heart rates higher than 65 beats per minute received either a single oral dose of 40 mg propranolol hydrochloride (Pranol; Dae Woong, Seoul, Korea) or a single dose (30 mg orally or 10 mg intravenously) of diltiazem hydrochloride (Herben; CJ, Seoul, Korea) 1 hour before the CT examination. The patients mean heart rate during the CT examination was 58 beats per minute 6 7 (range, beats per minute). Patients were scanned by using a 64-section CT scanner (Sensation 64; Siemens Healthcare, Forchheim, Germany). Before coronary CT angiography, a nonenhanced prospective ECGgated sequential scan was performed to measure CAC with the following parameters: rotation time, 330 msec; section collimation, 0.6 mm; section width, 3.0 mm; tube voltage, 120 kv; effective tube current time product, 50 mas; and table feed per scan, 18 mm. Thereafter in all patients, coronary CT angiography was performed by using retrospective ECG gating with ECG-based tube current modulation and the following parameters: rotation time, 330 msec; tube voltage, 120 kv; effective tube current time product, 800 mas; table feed per scan, 3.8 mm; and pitch, 0.2. A bolustracking technique (CARE bolus; Siemens Healthcare) was used for timing. As soon as attenuation of more than 100 HU was detected in the region of interest in the ascending aorta, the scan was initiated, with a delay of 6 seconds. Contrast enhancement was achieved with 75 ml iopamidol (370 mg iodine per milliliter, Iopamiro; Bracco, Milan, Italy) injected at 5 ml/sec, followed by an injection of 50 ml saline at 5 ml/sec by using a power injector (Envision CT; Medrad, Indianola, Pa) via the antecubital vein. The estimated volume CT dose index (CTDI vol ) (in milligrays) determined from the scanner s output parameters (peak kilovoltage and tube current time product per section) was recorded for each patient. The product of CTDI vol and scanning length (dose-length product in milligrays times centimeters) was calculated, and effective dose (in millisieverts) was estimated by using a normalization factor for the adult chest (0.017 msv mgy 2 1 cm 2 1 ) ( 13 ). Data Analysis CAC data and coronary CT angiograms were evaluated by using a workstation (Wizard; Siemens Healthcare). CAC was identified as a high-attenuation area in the coronary artery whose attenuation exceeded the threshold of 130 HU in a minimum of 3 contiguous pixels. CAC scores were calculated according to the Agatston method ( 14 ). For multivariate analysis, patients were grouped into five subsets on the basis of CAC score: those with CAC scores of 0, those with CAC scores of 1 10, those with CAC scores of , those with CAC scores of , and those with CAC scores greater than 400. Coronary CT angiograms were evaluated by two experienced cardiac radiologists (Y.J.K. and B.W.C., with 6 and 9 years of experience in cardiac CT, respectively) who were unaware of the clinical history of the patients. In case of disagreement, a joint reading was performed and a consensus was reached. Semiquantitative assessment was performed for all segments of the coronary artery tree, with an estimate of stenosis severity calculated as the ratio of the minimum contrast material enhanced lumen diameter to the diameter of a normal reference lumen of an unaffected distal segment. Obstructive CAD was defined as luminal narrowing of 50% or greater, whereas nonobstructive CAD was defined as less than 50% luminal narrowing. The extent of obstructive CAD was categorized into zero-, one-, two-, and three-vessel involvement. Left main artery involvement with obstructive CAD was considered as two-vessel involvement. Follow-up Patient follow-up data were collected by using electronic medical record review and/or standardized telephone interviews. The following events were regarded as clinical end points: cardiac death, nonfatal myocardial infarction (MI), un stable angina requiring hospitalization, and/or revascularization. Hard events were defined as cardiac death and nonfatal MI. Statistical Analysis All continuous data are expressed as means 6 standard deviations, and all categoric data are presented as percentages or absolute numbers. The Student t test and the x 2 test were used to assess differences between groups. The Cochran-Mantel-Haenszel test was used for analyzing the correlation between CAC scoring and coronary CT angiography. A composite MACE measure of cardiac death, nonfatal MI, unstable angina requiring hospitalization, and/or revascularization was used as an end point. Cumulative event rates as a function of time were obtained by using the Kaplan-Meier method for coronary CT angiography diagnosed CAD and were compared by using the log-rank test. Cox regression analyses were used to identify associations between clinical characteristics, multisection CT variables, and outcome. Univariate and multivariate analyses were performed to identify potential predictors. Hazard ratios (HRs) were calculated as an estimate of the risk associated with a particular variable, with 95% confidence intervals (CIs) based on binomial distributions. Receiver operating characteristic (ROC) curves were generated for four models: Model I, which included traditional risk factors; model II, which included CAC scores; model III, which included coronary CT angiography diagnosed CAD; and model IV, which included the combination of CAC scores and coronary CT angiography diagnosed CAD. Areas under the ROC curve (AUCs) were calculated to predict the ability of each model to detect the composite end point and were compared by the using DeLong method ( 15 ). Statistical analyses were performed by using software (SAS, version 9.1.3; SAS Institute, Cary, NC). P,.05 was considered to indicate a statistically significant difference. 94 radiology.rsna.org n Radiology: Volume 259: Number 1 April 2011

4 Table 1 Baseline Characteristics of Study Population according to Existence of CAD at Multisection CT Variable CAD Absent ( n = 2073) Present ( n = 1906) P Value Age (y) ,.001 No. of men 915 (44) 1146 (60),.001 Hypertension 873 (42) 1095 (57),.001 Diabetes 212 (10) 441 (23),.001 Current smoker 244 (12) 246 (13).523 BMI (kg/m 2 ) Dyslipidemia 722 (35) 808 (42),.001 Total cholesterol (mg/dl) Triglyceride level (mg/dl) ,.001 HDL cholesterol level (mg/dl) ,.001 LDL cholesterol level (mg/dl) Composite MACEs 2 (0.1) 103 (5.4),.001 CAC score * (52) 333 (8) (7) (18) (9) (5) CAD at coronary CT angiography * Normal fi ndings or 2073 (52) 1284 (32) nonobstructive CAD One-vessel disease (11) Two-vessel disease (4) Three-vessel disease (2) Note. Continuous values are means 6 standard deviations, and categoric values are numbers of patients, with percentages in parentheses. CAD was defi ned by the existence of any plaque at coronary CT angiography and/or the presence of CAC at CAC scoring. BMI = body mass index, HDL = high-density lipoprotein, LDL = low-density lipoprotein. To convert to Système International (SI) units (millimoles per liter), multiply total cholesterol, HDL cholesterol, and LDL cholesterol levels by and multiply triglyceride level by * Percentages are based on a denominator of 3979, the total number of patients examined. Table 2 Trend Analysis between CAC Score Subsets and Extent of CAD at Coronary CT Angiography CAC Score Category Normal Findings or Nonobstructive CAD One-Vessel CAD Two-Vessel CAD Three-Vessel CAD Total Note. Data are numbers of patients. P,.001 for relationship between increasing CAC score and increasing extent of CAD (shown by number of stenosed vessels at coronary CT angiography) (Cochran-Mantel-Haenszel test). Total Results Clinical Characteristics of the Cohort Overall, our study enrolled 4338 patients suspected of having CAD at presentation. From the overall study population, 359 patients with early elective revascularization were excluded from the analysis. After these patients were excluded, a total of 3979 patients (2061 [52%] men; mean age, 60 years 6 10; range, years) remained for further analysis. The distribution of the pretest likelihoods of CAD of this study cohort for low, intermediate, and high probability were 29% (1163 of 3979), 65% (2576 of 3979), and 6% (240 of 3979), respectively. Patient characteristics were categorized according to the existence of CAD. Patients with CAD tended to be older, male, hypertensive, diabetic, and dyslipidemic ( P,.001 for all) ( Table 1 ). CAC Scores and Coronary CT Angiography Results and Their Correlation The average CAC score in our cohort was (range, ), and the prevalence of CAC was 39% (1546 of 3979). The distribution according to CAC score subsets 0, 1 10, , , and greater than 400 was 60%, 7%, 18%, 9%, and 5%, respectively. The prevalence of obstructive CAD was 15% (622 of 3979), and the frequency according to the extent of obstructive CAD was 10%, 4%, and 2% for patients with one-, two-, and three-vessel obstructive CAD, respectively ( Table 1 ). Higher CAC scores were associated with a greater extent of CAD ( P,.001) ( Table 2 ). Radiation Dose The total estimated average radiation dose for the combined CAC scoring and coronary CT angiography acquisitions was 10.9 msv (range, msv). The average radiation dose for CAC scoring was 1.7 msv (range, msv), and the average radiation dose for coronary CT angiography only was 8.8 msv (range, msv). Follow-up Data Follow-up data were available in 99.9% (4332 of 4338) of patients, and the mean Radiology: Volume 259: Number 1 April 2011 n radiology.rsna.org 95

5 follow-up period for the study population was 828 days During follow-up, there were 105 composite MACEs, which consisted of cardiac death in two patients, nonfatal MI in three patients, unstable angina requiring hospitalization in one patient, and revascularization in 99 patients, for an event rate of 3% (105 of 3979). Importantly, there was a significant difference in the prevalence of composite MACEs according to the existence of CAD between the group with and that without CAD during the follow-up period (two of 2073 vs 103 of 1906, P,.001) ( Table 1 ). Although the prevalence of hard events was not significantly different between the group with and that without CAD according to the presence of CAD (one of 2073 vs four of 1906, P =.202), there was a significant statistical difference when we compared the prevalence of hard events according to the presence of obstructive CAD between the two groups (one of 3357 vs four of 622, P =.002). Prognostic Value Differentiated by Characteristics of CAD and CAC Score Subsets Event-free survival rates were 99%, 99%, 97%, 92%, and 83%, respectively, for patients in the CAC score subsets 0, 1 10, , , and greater than 400 ( P,.001) ( Fig 1a ). However, after adjustment for the traditional risk factors and the extent of CAD at coronary CT angiography, CAC score was not an independent predictor of MACEs ( Table 3 ). As for the prognostic value of the extent of CAD at coronary CT angiography, event-free survival rates were 99%, 90%, 80%, and 62% for patients with normal findings or nonobstructive CAD, one-vessel CAD, two-vessel CAD, and three-vessel CAD, respectively ( P,.001) ( Fig 1b ). Unadjusted and adjusted HRs for the composite MACE measure are shown in Table 3. The presence of obstructive CAD at coronary CT angiography had an independent prognostic predictive power for MACEs in the multivariate analysis adjusted for the traditional risk factors and CAC score. In addition, the prognostic predictive power for MACEs escalated according Table 3 HRs of Composite MACEs and 95% CIs according to CAC Score Subsets and Extent of CAD at Coronary CT Angiography Variable to the number of stenosed vessels, when compared with the group without obstructive CAD ( P,.001). AUCs for Predicting MACEs ROC curves were constructed to determine the predictive value of traditional risk factors, CAC scoring, the extent of CAD determined at coronary CT angiography, and CAC scoring in combination with coronary CT angiography for the composite MACE measure. The AUC was calculated for each model. In the cohort of this study, CAC scoring (model II) was more useful in the prediction of MACE than traditional risk factors (model I) ( P =.031). Furthermore, coronary CT angiography (model III) had better predictive value for MACEs than CAC scoring (model II) ( P,.001) and traditional risk factors (model I) ( P,.001). However, when CAC scoring was combined with coronary CT angiography (model IV), the resulting AUC demonstrated no significant difference in predictive power Univariate Analysis Multivariate Analysis HR 95% CI P Value HR 95% CIP Value Age , 1.09, , Male sex , , Hypertension , , Diabetes , 5.56, , Smoking , BMI , Dyslipidemia , , CAC score, , , , 7.92, , , 22.77, , , 47.98, , CAD at coronary CT,.001,.001 angiography One-vessel , 50.35, , 41.93,.001 disease Two-vessel , , , 88.07,.001 disease Three-vessel disease , , , ,.001 Note. HR adjustment included age, sex, hypertension, diabetes, dyslipidemia, CAC score, and extent of CAD at coronary CT angiography. BMI = body mass index. compared with the AUC for coronary CT angiography alone (model III) ( P =.198) ( Fig 2 ). Discussion The main finding in our study reveals that the extent of CAD diagnosed at coronary CT angiography has positive correlation with CAC scores and significantly predicts the risk of MACEs. Furthermore, coronary CT angiography has better predictive value than CAC scoring, and moreover, there is no added benefit to the addition of CAC scoring to coronary CT angiography in predicting MACEs in patients suspected of having CAD. CAC scoring has been used in clinical practice for risk stratification. Several studies ( 2 5 ) have demonstrated that higher CAC scores are associated with increased plaque burden and increased cardiovascular risk. On the other hand, coronary CT angiography provides comprehensive information 96 radiology.rsna.org n Radiology: Volume 259: Number 1 April 2011

6 Figure 1 Figure 1: Adjusted survival probability plots show event-free survival. (a) Plot of survival according to CAC scoring results shows that even after adjustment, the event-free survival rate is proportionally worse as the CAC score increases. The P values were.666,.001, less than.001, and less than.001, respectively, for CAC scores of 1 10, , , and greater than 400 when compared with a CAC score of 0. (b) Plot of survival according to coronary CT angiography results shows that even after adjustment, event-free survival rate is proportionally worse as the extent of CAD increases. The P values were all less than.001 for one-, two-, and three-vessel obstructive CAD when compared with normal fi ndings or nonobstructive CAD. Adjustments were performed for traditional risk factors, including age, sex, hypertension, diabetes, and dyslipidemia. Survival times are expressed in days. re garding the location and severity of atherosclerotic plaque in the coronary arteries. Several studies have taken advantage of this feature of coronary CT angiography and demonstrated independent prognostic utility of coronary CT angiography in patients suspected of having CAD ( 11,16 18 ). Interestingly, a positive association has been observed between CAC scores and both the presence and extent of significant CAD ( 16,19 ). In accordance with previous studies, our study demonstrates that the extent of CAD has positive correlation with CAC scores. Furthermore, the extent of CAD was an independent predictor of MACEs in patients suspected of having CAD. However, as our study population was a relatively low-risk population with no prior documented CAD, the hard event rate was low at the current duration of follow-up. The power calculation for the sample size was performed for models III and IV by using PASS (Number Cruncher Statistical System statistical software, 2008). The power for the sample size of our study was.1704 (17.04%). The primary reason for such relatively low power of the sample size was the low prevalence of MACEs (, 3%) in our study population. Up to the present, it has been customary to perform both CAC scoring and coronary CT angiography for diagnosis in patients suspected of having CAD. Results dealing with the incremental prognostic value of CAC scoring used in combination with coronary CT angiography have recently been published ( 11 ). Hence, in the electron-beam CT era, combined CAC scoring and coronary CT angiography was thought to be the optimal protocol for cardiac CT in predicting cardiovascular events. However, contemporary multisection CT technology provides better spatial and temporal resolution compared with those of electron-beam CT, thereby offering improved diagnostic accuracy for CAD ( 8 10,20 23 ). To our knowledge, the relative value of coronary CT angiography and CAC scoring for prognostication has not previously been studied. It is relevant to assess the prognostic implications of coronary CT angiography and CAC scoring for potentially reducing radiation exposure and streamlining CT protocols. Our results demonstrated that the prognostic value of coronary CT angiography is significantly greater than that of CAC scoring. Most importantly, the additive prognostic value of CAC scoring to coronary CT angiography was shown to be nonsignificant. Although the added value of CAC scoring for prognosis may be limited, it is well known that a very high CAC score (. 1000) may affect the diagnostic performance of coronary CT angiography and may lead to overestimation of the degree of stenosis. In particular, because of the very high prevalence of significant stenosis and the decreased diagnostic accuracy of coronary CT angiography in patients with very high CAC scores, further evaluation with coronary CT angiography may not be indicated, as these patients would probably be better served by undergoing invasive coronary Radiology: Volume 259: Number 1 April 2011 n radiology.rsna.org 97

7 Figure 2 angiography ( 24 ). Our conclusions may not apply to patients with very high calcium scores. We examined a limited number of patients with high calcium scores, precluding definite conclusions for these particular patients. The basic principle of radiation protection is to keep radiation exposure as low as reasonably achievable (the ALARA principle) ( 25 ). According to this principle, if CAC scoring has no added benefit over coronary CT angiography, considering radiation hazards, a combined CAC scoring and coronary CT angiography scanning protocol confers no benefit and contains harm for individuals undergoing cardiac CT. Still, radiation exposure from CAC scoring is relatively small compared with that from coronary CT angiography in contemporary multisection CT technology, so radiation by coronary CT angiography poses a greater threat to the patient than performing CAC scoring. However, given the fast improvements in dose reduction in coronary CT angiography Figure 2: ROC curves of four models for predicting the composite MACE measure. Model I incorporated traditional risk factors, including age, sex, hypertension, diabetes, and dyslipidemia; model II, CAC scoring; model III, coronary CT angiography; and model IV, the combination of CAC scoring and coronary CT angiography. P =.031 for model II versus model I, P,.001 for model III versus model II, and P =.198 for model IV versus model III. SE = standard error of the estimate. scanning, the radiation hazards between the two modalities will be comparable in the near future. Several limitations of our study should be acknowledged. First, our study was performed at a single center, which makes it uncertain whether results will be equally applicable to general clinical practice. Second, treatment bias is possible, in that patients with significant CAD are more likely to be treated aggressively, resulting in increased revascularization, which constituted the major proportion of the composite MACE measure. To avoid the treatment bias, patients who underwent elective revascularization within 60 days after the index CT study were excluded ( 16 ). Moreover, patients underwent multisection CT for symptom evaluation, and all decisions regarding revascularization were based on symptoms and/or the presence of concomitant ischemia at noninvasive testing followed by invasive coronary angiography ( 18 ). Besides, the presence of obstructive CAD at multisection CT had a significant association with hard cardiac events. As previously mentioned, a longer follow-up is clearly warranted to confirm these results. Third, the calculated power for the sample size was relatively low because of the low prevalence of MACEs in our study population. Therefore, a larger cohort study may be needed for verification. In the multisection CT era, coronary CT angiography is better than CAC scoring in predicting MACEs in low-risk patients suspected of having CAD. Furthermore, the current standard multisection CT protocol (coronary CT angiography combined with CAC scoring) has no incremental prognostic value over coronary CT angiography alone in such individuals. Therefore, in terms of prognosis, CAC scoring may no longer need to be incorporated into the cardiac CT protocol in this population, considering the radiation exposure. Disclosures of Potential Conflicts of Interest: S.W.K. No potential conflicts of interest to disclose. Y.J.K. No potential conflicts of interest to disclose. J.S. No potential conflicts of interest to disclose. J.M.S. No potential conflicts of interest to disclose. M.E.H. No potential conflicts of interest to disclose. D.W.K. No potential conflicts of interest to disclose. J.Y.K. No potential conflicts of interest to disclose. B.W.C. No potential conflicts of interest to disclose. H.J.C. No potential conflicts of interest to disclose. References 1. Meijboom WB, van Mieghem CA, Mollet NR, et al. 64-slice computed tomography coronary angiography in patients with high, intermediate, or low pretest probability of significant coronary artery disease. J Am Coll Cardiol 2007 ; 50 ( 15 ): Detrano R, Guerci AD, Carr JJ, et al. Coronary calcium as a predictor of coronary events in four racial or ethnic groups. N Engl J Med 2008 ; 358 ( 13 ): Budoff MJ, Shaw LJ, Liu ST, et al. Longterm prognosis associated with coronary calcification: observations from a registry of 25,253 patients. J Am Coll Cardiol 2007 ; 49 ( 18 ): Shaw LJ, Raggi P, Schisterman E, Berman DS, Callister TQ. Prognostic value of cardiac risk factors and coronary artery calcium screening for all-cause mortality. Radiology 2003 ; 228 ( 3 ): Nasir K, Shaw LJ, Liu ST, et al. Ethnic differences in the prognostic value of coronary 98 radiology.rsna.org n Radiology: Volume 259: Number 1 April 2011

8 artery calcification for all-cause mortality. J Am Coll Cardiol 2007 ; 50 ( 10 ): Scholte AJ, Schuijf JD, Kharagjitsingh AV, et al. Prevalence of coronary artery disease and plaque morphology assessed by multislice computed tomography coronary angiography and calcium scoring in asymptomatic patients with type 2 diabetes. Heart 2008 ; 94 ( 3 ): Hoffmann MH, Shi H, Schmitz BL, et al. Noninvasive coronary angiography with multislice computed tomography. JAMA 2005 ; 293 ( 20 ): Miller JM, Rochitte CE, Dewey M, et al. Diagnostic performance of coronary angiography by 64-row CT. N Engl J Med 2008 ; 359 ( 22 ): Budoff MJ, Dowe D, Jollis JG, et al. Diagnostic performance of 64-multidetector row coronary computed tomographic angiography for evaluation of coronary artery stenosis in individuals without known coronary artery disease: results from the prospective multicenter ACCURACY (Assessment by Coronary Computed Tomographic Angiography of Individuals Undergoing Invasive Coronary Angiography) trial. J Am Coll Cardiol 2008 ; 52 ( 21 ): Meijboom WB, Meijs MF, Schuijf JD, et al. Diagnostic accuracy of 64-slice computed tomography coronary angiography: a prospective, multicenter, multivendor study. J Am Coll Cardiol 2008 ; 52 ( 25 ): Ostrom MP, Gopal A, Ahmadi N, et al. Mortality incidence and the severity of coronary atherosclerosis assessed by computed tomography angiography. J Am Coll Cardiol 2008 ; 52 ( 16 ): Gibbons RJ, Balady GJ, Bricker JT, et al. ACC/AHA 2002 guideline update for exercise testing: summary article a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1997 Exercise Testing Guidelines). J Am Coll Cardiol 2002 ; 40 ( 8 ): Bongartz G, Golding SJ, Jurik AG, et al. European guidelines on quality criteria for computed tomography. Publication EUR EN. /htmlindex.htm. Accessed February 20, Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M Jr, Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol 1990 ; 15 ( 4 ): DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 1988 ; 44 ( 3 ): van Werkhoven JM, Schuijf JD, Gaemperli O, et al. Incremental prognostic value of multislice computed tomography coronary angiography over coronary artery calcium scoring in patients with suspected coronary artery disease. Eur Heart J 2009 ; 30 ( 21 ): Min JK, Shaw LJ, Devereux RB, et al. Prognostic value of multidetector coronary computed tomographic angiography for prediction of all-cause mortality. J Am Coll Cardiol 2007 ; 50 ( 12 ): Carrigan TP, Nair D, Schoenhagen P, et al. Prognostic utility of 64-slice computed tomography in patients with suspected but no documented coronary artery disease. Eur Heart J 2009 ; 30 ( 3 ): Ho JS, Fitzgerald SJ, Stolfus LL, et al. Relation of a coronary artery calcium score higher than 400 to coronary stenoses detected using multidetector computed tomography and to traditional cardiovascular risk factors. Am J Cardiol 2008 ; 101 ( 10 ): Hamon M, Morello R, Riddell JW, Hamon M. Coronary arteries: diagnostic performance of 16- versus 64-section spiral CT compared with invasive coronary angiography metaanalysis. Radiology 2007 ; 245 ( 3 ): Hausleiter J, Meyer T, Hadamitzky M, et al. Non-invasive coronary computed tomographic angiography for patients with suspected coronary artery disease: the Coronary Angiography by Computed Tomography with the Use of a Submillimeter resolution (CACTUS) trial. Eur Heart J 2007 ; 28 ( 24 ): Scheffel H, Alkadhi H, Plass A, et al. Accuracy of dual-source CT coronary angiography: first experience in a high pre-test probability population without heart rate control. Eur Radiol 2006 ; 16 ( 12 ): Dewey M, Zimmermann E, Deissenrieder F, et al. Noninvasive coronary angiography by 320-row computed tomography with lower radiation exposure and maintained diagnostic accuracy: comparison of results with cardiac catheterization in a head-to-head pilot investigation. Circulation 2009 ; 120 ( 10 ): Palumbo AA, Maffei E, Martini C, et al. Coronary calcium score as gatekeeper for 64- slice computed tomography coronary angiography in patients with chest pain: per-segment and per-patient analysis. Eur Radiol 2009 ; 19 ( 9 ): U.S. Nuclear Regulatory Commission. 10 CFR Definitions. /reading-rm/doc-collections/cfr/part020 /part html. Accessed January 12, Radiology: Volume 259: Number 1 April 2011 n radiology.rsna.org 99