Introduction. Aims. Keywords coronary artery disease coronary computed tomographic angiography intravascular ultrasound

Transcription

1 European Heart Journal Cardiovascular Imaging (2015) 16, doi: /ehjci/jev100 Limitations of coronary computed tomographic angiography for delineating the lumen and vessel contours of coronary arteries in patients with stable angina Choongki Kim 1, Sung-Jin Hong 1, Dong-Ho Shin 1,2, Jung-Sun Kim 1,2, Byeong-Keuk Kim 1,2, Young-Guk Ko 1,2, Donghoon Choi 1,2, Yangsoo Jang 1,2,3, and Myeong-Ki Hong 1,2,3 * 1 Severance Cardiovascular Hospital, Yonsei University Health System, Seoul, Korea; 2 Cardiovascular Research Institute, Yonsei University College of Medicine, Seoul, Korea; and 3 Severance Biomedical Science Institute, Yonsei University College of Medicine, Seoul, Korea Received 6 January 2015; accepted after revision 31 March 2015; online publish-ahead-of-print 30 April 2015 Aims We sought to evaluate whether coronary computed tomographic angiography (CCTA) could accurately and reproducibly delineate the lumen and vessel contours of coronary arteries.... Methods One hundred coronary stenotic lesions representing 91 patients with stable angina who received both CCTA and intravascular ultrasound (IVUS) were analysed. Three segments with minimal lumen cross-sectional area (CSA), proximal ref- and results erence, and distal reference on IVUS images were selected for each lesion. Five observers measured lumen and vessel CSAs at three matching segments on CCTA images. These CSAs were compared with the IVUS-measured CSAs as a reference standard. All five observers underestimated lumen CSA at the three selected segments by CCTA. The minimal lumen CSA assessed by CCTA exhibited very weak correlations with those obtained by IVUS (r ¼0.23, 0.24, 0.15, 0.25, and 0.28, respectively). In contrast to the lumen CSA, the vessel CSA at the three segments was overestimated by all observers when assessed by CCTA. At the segment with minimal lumen CSA, the vessel CSA obtained by CCTA showed weak correlations with those assessed by IVUS (r ¼ 0.43, 0.33, 0.44, 0.37, and 0.42, respectively). Moreover, intra-class correlation coefficients ranged from 0.44 to 0.73 among the five observers for lumen or vessel CSA measurements by CCTA at the segment with minimal lumen CSA.... Conclusion CCTA has potential limitations in the accurate delineation of lumen and vessel contours in patients with angina, as there was a high level of discordance with the IVUS-measured lumen and vessel CSAs and high inter-observer variability Keywords coronary artery disease coronary computed tomographic angiography intravascular ultrasound Introduction Coronary computed tomographic angiography (CCTA) is believed to be a dependable non-invasive method for screening or diagnosing coronary artery disease. 1 3 CCTA allows the haemodynamics of coronary artery stenosis to be assessed through accurate delineation of stenotic lesions with significantly narrow lumens. 4 However, the ability of CCTA to delineate the lumen or vessel border has not been fully evaluated in patients with stable angina. Such an approach would be valuable for screening asymptomatic patients for the presence of coronary artery disease. Although automated quantitative CCTA algorithms have improved reproducibility and time efficiency with contouring of the coronary artery, the accuracy of many algorithms requires manual centre lines to be annotated. Therefore, in the present study, we sought to evaluate whether CCTA could precisely delineate the lumen and vessel borders compared with * Corresponding author. Division of Cardiology, Severance Cardiovascular Hospital, Yonsei University College of Medicine, 250 Seongsanno, Seodaemun-gu, Seoul , Republic of Korea. Tel: ; Fax: , mkhong61@yuhs.ac The authors contributed equally to this study. Published on behalf of the European Society of Cardiology. All rights reserved. & The Author For permissions please journals.permissions@oup.com.

2 Limitations of CCTA 1359 intravascular ultrasound (IVUS), which is used as the gold standard with accuracy and reproducibility for evaluation of coronary lumen and vessel contour, 5 7 particularly in patients with stable angina who received coronary angiograms in the cardiac catheterization laboratory in routine clinical practice. Methods We examined the IVUS database at our institute and retrospectively identified 100 coronary stenotic lesions among 91 patients with stable angina who received CCTA before coronary angiography and IVUS. Patients who had prior coronary artery bypass surgery or stent implantation were excluded. Patients with atrial fibrillation, acute coronary syndromes, totally occluded lesions or poor CCTA images were also excluded. IVUS was performed in a standard fashion with a 40-MHz IVUS catheter (Atlantis, Boston Scientific Corp/SCIMED, Minneapolis, MN, USA) before any intervention and after intracoronary administration of 0.2 mg nitroglycerin through an automatic motorized transducer pullback system (0.5 mm/s). Quantitative IVUS analyses were performed by a single experienced cardiologist who was blinded to the CCTA measurements and the patients clinical and procedural data according to the criteria of the American College of Cardiology Clinical Expert Consensus Document on IVUS. 7 These analyses were performed with computerized planimetry (EchoPlaque, Indec Systems Inc., Santa Clara, CA, USA) at the segment with minimal lumen cross-sectional area (CSA) and proximal and distal reference segments. The reference segments were the most normal-looking cross-sections within 10-mm proximal and distal locations to the lesion but before any side branch. Quantitative measurements included external elastic membrane and lumen CSAs. In the presence of more than one coronary lesion per vessel, the more severe lesion was analysed. All patients received CCTA examinations within 14 days prior to conventional coronary angiography. CCTA imaging was performed with a 64-channel CT scanner platform (Somatom Sensation and Definition CT, Siemens, Forchheim, Germany) using standardized protocols for image acquisition as defined by the Society of Cardiovascular Computed Tomography. 8 Oral metoprolol was administered 1 or 2 h prior to CCTA examination if the patient had a baseline heart rate of 70 beats/min and no contraindications. A non-enhanced prospective electrocardiographictriggered scan was performed prior to the contrast-enhanced helical scan to score calcium levels. Scanning parameters were as follows: 120 kvp with mas (depending on the size of the patient), mm slice collimation, and 330-ms gantry rotation time. Sublingual nitroglycerin (0.2 mg) was administered immediately before contrast injection. A contrast-enhanced CCTA was performed using retrospective electrocardiography-triggered acquisitions with the following scan parameters: kvp, mas (depending on the size of the patient), mm slice collimation, and 330-ms gantry rotation time. A real-time bolus-tracking method was used to trigger the initiation of the scan. A bolus of ml of iopamidol was injected into an antecubital vein at a flow rate of 5 ml/s, followed by a 50-mL saline flush at a flow rate of 5 ml/s. The scans were started 7 s after reaching the threshold of 140 Hounsfield units. In the cross-sectional images perpendicular to the vessel centreline determined for each major coronary artery with the use of dedicated offline software (Vitrea 2.0, Vital Images, Minnetonka, MN, USA), contours of lumens and vessels were manually traced on a magnified view of the multiplanar reformatted images. The external elastic membrane and lumen contours were traced at the segment with minimal lumen CSA and proximal and distal references on both IVUS and CCTA cross-sectional images. For accurate comparisons of IVUS and CCTA, fiduciary points, such as side branches or characteristic calcifications, were selected to match corresponding coronary segments. The distance from the coronary ostium to each fiduciary point was measured using the longitudinal reconstructed IVUS data sets and multiplanar reconstructions of the CCTA datasets to ensure that distance measurements were the same with both methods. All matching between CCTA and IVUS images was done by an independent observer. We confirmed that the distance between IVUS and CCTA was the same for all 100 lesions. CCTA images were independently analysed by five observers (designated A, B, C, D, and E) who were blinded to the patients identities and were highly trained for CCTA image analysis with sufficient experience for.5 years. The five observers traced the lumen and vessel contours manually at the three matching segments with minimal lumen CSA, proximal reference, and distal reference and measured the lumen and vessel CSAs on CCTA. These CCTA-measured CSAs were compared with IVUS-measured CSAs as the reference standard. 5 7 The consistency of CCTA-measured CSAs was also evaluated for the five observers. Additionally, significant plaques were classified as calcified when they mainly contained high-density components ( 130 Hounsfield units); as non-calcified when they were composed of low-density components; and as mixed when they contained both high- and low-density components. Representative cases are shown for measurements of lumen and vessel CSAs by IVUS and CCTA images for a calcified lesion (Figure 1) and for a non-calcified lesion (Figure 2). Continuous data were presented as means + standard deviations or median (25th, 75th percentile), and categorical data were reported as frequencies. One-way analysis of variance or Student s t-test was used for comparison of continuous variables. The correlation and difference between IVUS measurements and each of the five observer s CCTA measurements were assessed by the Pearson correlation coefficient and Bland Altman method. Inter-observer and intra-observer s were measured by intra-class correlation coefficient. Statistical analyses were performed with SPSS version 18.0 (SPSS Inc., Chicago, IL, USA). A P-value of,0.05 was considered significant. Results Baseline characteristics of the study participants are presented in Table 1. Mean lumen and vessel CSA measurements of 100 stenotic lesions obtained by CCTA by five observers and IVUS are presented in Table 2. All five observers underestimated the lumen CSAs at the segments with minimal lumen CSAs, proximal reference segments, and distal reference segments with CCTA compared with those assessed by IVUS (Table 2). The lumen CSAs measured by the five observers showed weak to moderate correlations with IVUSmeasured CSAs (Table 3). In particular, the Pearson correlation coefficient between IVUS- and CCTA-measured lumen CSAs was only 0.23, 0.24, 0.15, 0.25, and 0.28, respectively, in the segment with minimal lumen CSA. The 95% limits of s between IVUS- and CCTA-measured lumen CSAs ranged from +3.7 to +4.7 mm 2 for absolute CSA difference (Figure 3). Consequently, the proportions of the lesions that agreed with the margin of error of,0.5 mm 2 for lumen CSA at the segment with minimal lumen CSA were only 14, 29, 22, 19, and 23%, according to the five observers. In contrast to lumen CSA measurements, all five observers overestimated vessel CSAs of the three segments by CCTA (Table 2). The CCTA-measured vessel CSAs by five observers were moderately correlated with IVUS-derived vessel CSAs at the three segments (Table 3). The Pearson correlation coefficients

3 1360 C. Kim et al. Figure 1 Representative case of measurements of lumen and vessel CSAs at the proximal reference segment, the segment with minimal lumen CSA, and the distal reference segment by IVUS and CCTA for a calcified lesion. Five observers (A E) traced the lumens and vessels on CCTA images. between IVUS- and CCTA-derived vessel CSAs were 0.43, 0.33, 0.44, 0.37, and 0.42, respectively, in the segment with minimal lumen CSA. The 95% limits of s between IVUS- and CCTA-measured vessel CSAs ranged from +9.0 to mm 2 for absolute CSA difference (Figure 4). Consequently, the proportions of the lesions that agreed with the margin of error of,1.0 mm 2 of vessel CSAs at the segment with minimal lumen CSA were only 22, 15, 16, 15, and 22 according to the five observers. An observer A repeated these measurements1week after the initial CCTA measurements. Intra-class correlation coefficients for intraobserver variability were 0.79 for lumen CSA and 0.67 for vessel CSA at the segment with minimal lumen CSA. Inter-observer reliabilities between each pair of measurements by the five observers are presented in Table 4. Intra-class correlation coefficients ranged from 0.59 to 0.73 for lumen CSA in the segment with minimal lumen CSA. Regarding the vessel CSA measurements at the segment with minimal lumen CSA, intra-class correlation coefficients ranged from 0.44 to 0.70 (Table 4). Absolute differences of lumen and vessel CSA between IVUS and CCTA measurements at the segment with minimal lumen CSA are shown according to various subsets (Figure 5). Discussion The principal findings of the present study were as follows. (i) All five observers underestimated lumen CSAs and overestimated vessel CSAs at the matching three segments by CCTA compared with IVUS. (ii) Each CCTA-measured lumen or vessel CSAs of the five observers showed weak correlations with IVUS-measured CSAs, particularly at the segment with minimal lumen CSA. (iii) Only a moderate degree of inter-observer reliability was observed among the five observers; the reliabilities decreased as the lumen CSAs were measured at the segment with smaller vessel diameters. CCTA has been used for non-invasive, imaging-based diagnosis of coronary artery disease. Recent investigations attempted to accurately visualize anatomic lesions, including plaques, and to determine the haemodynamic significance of coronary artery disease. 4,9,10 Previous studies reported that CCTA measurements showed good correlations with IVUS measurements, including good inter-observer and intra-observer variabilities for coronary stenosis; however, variability was acceptable for the presence of coronary stenosis but was not validated for precise quantitative measurements. Despite the development of image processing protocols and CCTA contour detection algorithms, the limitations of this method remain a major challenge. In the present study, CCTA showed a smaller lumen CSA and greater vessel CSA compared with those of IVUS. Therefore, CCTA measurements were considered to overestimate the severity of coronary artery stenosis. One study with 57 patients who presented chest pain at an outpatient clinic showed that minimal lumen CSA was underestimated by CCTA with 95% limits of

4 Limitations of CCTA 1361 Figure 2 Representative case of measurements of lumen and vessel CSAs at the proximal reference segment, the segment with minimal lumen CSA, and the distal reference segment by IVUS and CCTA for a non-calcified lesion. Five observers (A E) traced the lumens and vessels on CCTA images. Table 1 Baseline characteristics of study population Variables Total 91 patients (Total 100 lesions).. Age (years) Male 62 (68%) Body mass index (kg/m 2 ) Hypertension 48 (53%) Diabetes mellitus 30 (33%) Current smoker 32 (35%) Dyslipidaemia 15 (16%) Total coronary calcium score (Agatston units) 147 (43 472) Calcified or mixed plaque 38 (38%) Lesion vessel Left main 1 (1%) Left anterior descending artery 74 (74%) Left circumflex artery 10 (10%) Right coronary artery 15 (15%) Heart rate during computed tomography 58 (54 64) scanning (beats/min) Data are presented as frequencies (%), means + SDs, or median (25th, 75th percentile). for a bias ranging from to 2.30 mm 2 ; median minimal lumen CSA on CCTA was 4.3 mm 2 compared with 5.2 mm 2 on IVUS. 10 However, another study with 60 patients (angina in 50% and shortness of breath in 28%) reported that minimal lumen CSA was overestimated on CCTA by 21%, and mean minimal lumen CSA on CCTA was 5.5 mm 2 compared with 4.1 mm 2 on IVUS; the limits of between IVUS and CCTA measurement was +3.8 mm 2 for the absolute CSA difference. 13 A recent meta-analysis showed that CCTA slightly overestimates lumen CSA compared with that of IVUS. 14 However, each of the studies included in that meta-analysis showed wide and varied mean differences and confidence intervals of for lumen CSAs by CCTA and IVUS. The mean difference ranged from to 1.40 mm 2, and the 95% confidence interval ranged from to mm The limitation of a small sample size and different methodologies and characteristics of studies, including scanners, clinical presentation of patients, proportions of plaque characteristics (calcified vs. non-calcified plaques), different degrees of lesion severity (severely vs. moderately or mildly narrowed lesions), and other study protocols may have affected the different outcomes of each study In addition, the different methods of nitroglycerin administration (direct intracoronary vs. sublingual) between IVUS and CCTA might elicit different outcomes. Previous studies reported that sublingual nitroglycerin administration improves lumen diameter

5 1362 C. Kim et al. Table 2 Lumen and vessel cross-sectional area obtained by IVUS and five CCTA observers IVUS-measured CCTA-measured CSA, mm 2 P-value b CSA, mm 2... Observer A Observer B Observer C Observer D Observer E Mean P-value a... Lumen CSA Proximal reference ,0.001 Minimal lumen CSA ,0.001 Distal reference ,0.001 Vessel CSA Proximal reference Minimal lumen CSA ,0.001 Distal reference Data are presented as means + SDs. CCTA, coronary computed tomographic angiography; CSA, cross-sectional area; IVUS, intravascular ultrasound. a One-way ANOVA between five observers measurements. b IVUS-measured and mean CCTA-measured CSA by five observers. Table 3 Correlation and limit of between IVUS- and CCTA-measured cross-sectional areas according to five observers Observer A Observer B Observer C Observer D Observer E r 95% limit of r 95% limit of r 95% limit of r 95% limit of r 95% limit of... Lumen CSA Proximal reference Minimal lumen CSA Distal reference Vessel CSA Proximal reference Minimal lumen CSA Distal reference The 95% limit of was proposed with 1.96 standard deviations of difference between IVUS and CCTA measurements. CCTA, coronary computed tomographic angiography; CSA, cross-sectional area; IVUS, intravascular ultrasound. measurements and vascular visualization in CCTA measurements. 17,18 Sublingual nitroglycerin was administered immediately before contrast injection in the CCTA examination of this study. Intracoronary nitroglycerin was also administered immediately before IVUS examination in this study. Based on the different degree of coronary artery dilation depending on the route of nitroglycerin administration (direct intracoronary vs. sublingual), a greater degree of coronaryartery dilation may be expectedin cases with direct intracoronary administration of nitroglycerin compared with those with sublingual administration. Subsequently, different methods of nitroglycerin administration may be important factors that cause underestimation of lumen CSA by CCTA. Our findings demonstrated poor correlations between CCTA and IVUS for quantitative measurements of lumen and vessel CSAs, especially at the segment with minimal lumen CSA. A prospective comparison of IVUS and CCTA for quantitative measurement of minimal lumen CSA in intermediate lesions showed a moderate correlation coefficient (r ¼ 0.65) and wide limits of (+3.81 mm 2 for absolute lumen CSA difference). 13 The previous study suggested that quantitative CCTA lumen measurements might be safely used on a population basis, whereas the predictive accuracy of the individual patient is limited. 13 Thus, measuring lumen CSA by CCTA may not be a suitable method for replacing IVUS as a precise quantitative test in patients with stable angina. Similarly, the results of the present study showed a moderate degree of inter-observer reliability among five observers and wide limits of. Moreover, lower inter-observer reliability of lumen CSA measurement was observed in lesions with smaller reference diameters. These findings suggest that poor border detection or limitations of manual tracing on CCTA cross-sectional images due to

6 Limitations of CCTA 1363 Figure 3 Correlation and Bland Altman plots for IVUS measurements and CCTA measurements of lumen CSAs at the segment with minimal lumen CSA by five observers. Upper panel: Measurements of minimal lumen CSAs by CCTA showed poor correlations with IVUS-measured minimal lumen CSAs for 100 coronary stenotic lesions. Blue lines indicate the margin of error at 0.5 mm 2 for CCTA measurements compared with IVUS measurements as the reference standard. The proportions within the margins of error were 14, 29, 22, 19, and 23% for observers A E, respectively. Lower panel: Bland Altman plots. Solid red lines indicate bias (mean difference), and hatched red lines indicate 95% limits of. Figure 4 Correlation and Bland Altman plots for IVUS measurements and CCTA measurements of vessel CSAs at the segment with minimal lumen CSA by five observers. Upper panel: Measurements of vessel CSAs by CCTA showed moderate correlations with IVUS-measured vessel CSAs for 100 coronary stenotic lesions. Blue lines indicate the margin of error at 1.0 mm 2 for CCTA measurements compared with IVUS measurements as the reference standard. The proportions within the margins of error were 22, 15, 16, 15, and 22%, respectively, for observers A E. Lower panel: Bland Altman plots. Solid red lines indicate bias (mean difference), and hatched red lines indicate 95% limits of. relatively low resolution may be accentuated in the small distal segment of the coronary artery. Similar to the results of the present study, previous studies reported that lower reliability of plaque measurements and detection of significant stenosis by CCTA was more prominent at small distal segments and calcified lesions Improvements in CCTA technologies (i.e. enhanced resolution for relatively poor border detection and special methods for calcified lesions or small-sized vessel) will be required for accurate measurements of coronary lumen and vessel contour in patients with stable angina.

7 1364 C. Kim et al. Table 4 Inter-observer reliability among five observers Variables Observer Intra-class correlation coefficient (95% confidence interval)... Observer B Observer C Observer D Observer E... Lumen CSA Proximal reference A 0.80 ( ) 0.85 ( ) 0.79 ( ) 0.87 ( ) B 0.77 ( ) 0.69 ( ) 0.77 ( ) C 0.73 ( ) 0.79 ( ) D 0.75 ( ) Minimal lumen CSA A 0.61 ( ) 0.67 ( ) 0.67 ( ) 0.73 ( ) B 0.65 ( ) 0.69 ( ) 0.67 ( ) C 0.67 ( ) 0.59 ( ) D 0.61 ( ) Distal reference A 0.81 ( ) 0.78 ( ) 0.70 ( ) 0.80 ( ) B 0.85 ( ) 0.76 ( ) 0.82 ( ) C 0.69 ( ) 0.77 ( ) D 0.68 ( ) Vessel CSA Proximal reference A 0.62 ( ) 0.83 ( ) 0.72 ( ) 0.82 ( ) B 0.72 ( ) 0.72 ( ) 0.76 ( ) C 0.75 ( ) 0.83 ( ) D 0.79 ( ) Minimal lumen CSA A 0.45 ( ) 0.67 ( ) 0.44 ( ) 0.60 ( ) B 0.57 ( ) 0.65 ( ) 0.67 ( ) C 0.55 ( ) 0.70 ( ) D 0.63 ( ) Distal reference A 0.72 ( ) 0.76 ( ) 0.71 ( ) 0.79 ( ) B 0.72 ( ) 0.74 ( ) 0.81 ( ) C 0.59 ( ) 0.78 ( ) D 0.70 ( ) CSA, cross-sectional area. Figure 5 Comparison between IVUS-measured CSA and CCTA-measured CSA at the segment with minimal lumen CSA according to various subsets. Upper panels show absolute difference of lumen CSA between IVUS and CCTA measurements. Lower panels show absolute difference of vessel CSA between IVUS and CCTA measurements. RD, reference vessel diameter, HR, heart rate.

8 Limitations of CCTA 1365 A potential limitation was that this study was based on retrospective analysis of patients at a single centre. Use of IVUS during conventional angiography was determined at the discretion of the operator. Finally, the patients with poor image quality of CCTA were excluded from the analyses, which can worsen the accuracy and reproducibility. In conclusion, CCTA has potential limitations for delineating the lumen or vessel contours. High discordance was found with IVUSderived lumen or vessel CSAs, and high inter-observer variability existed for measurement of lumen or vessel CSAs. Conflict of interest: None declared. Funding This study was supported by a grant from the Korea Healthcare Technology R&D Project, Ministry for Health, Welfare & Family Affairs, Republic of Korea (Nos. A and A102064) and the Cardiovascular Research Center, Seoul, Korea. References 1. Achenbach S, Moselewski F, Ropers D, Ferencik M, Hoffmann U, MacNeill B et al. Detection of calcified and noncalcified coronary atherosclerotic plaque by contrast-enhanced, submillimeter multidetector spiral computed tomography: a segment-based comparison with intravascular ultrasound. Circulation 2004;109: Leber AW, KnezA, von Ziegler F, BeckerA, Nikolaou K, PaulS et al. Quantificationof obstructiveandnonobstructivecoronarylesions by64-slice computedtomography: a comparative study with quantitative coronary angiography and intravascular ultrasound. J Am Coll Cardiol 2005;46: Budoff MJ, Dowe D, Jollis JG, Gitter M, Sutherland J, Halamert E 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: Min JK, Leipsic J, Pencina MJ, Berman DS, Koo BK, van Mieghem C et al. Diagnostic accuracy of fractional flow reserve from anatomic CT angiography. JAMA 2012; 308: Peters RJ, Kok WE, Rijsterborgh H, van Dijk M, Koch KT, Piek JJ et al. Reproducibility of quantitative measurements from intracoronary ultrasound images. Beat-to-beat variability and influence of the cardiac cycle. Eur Heart J 1996;17: von Birgelen C, de Vrey EA, Mintz GS, Nicosia A, Bruining N, Li W et al. ECG-gated three-dimensional intravascular ultrasound: feasibility and reproducibility of the automated analysis of coronary lumen and atherosclerotic plaque dimensions in humans. Circulation 1997;96: Mintz GS, Nissen SE, Anderson WD, Bailey SR, Erbel R, Fitzgerald PJ et al. American College of Cardiology Clinical Expert Consensus Document on Standards for Acquisition, Measurement and Reporting of Intravascular Ultrasound Studies (IVUS). A report of the American College of Cardiology Task Force on Clinical Expert Consensus Documents. J Am Coll Cardiol 2001;37: Taylor AJ, Cerqueira M, Hodgson JM, Mark D, Min J, O Gara P et al. ACCF/SCCT/ ACR/AHA/ASE/ASNC/NASCI/SCAI/SCMR 2010 appropriate use criteria for cardiac computed tomography. A report of the American College of Cardiology Foundation Appropriate Use Criteria Task Force, the Society of Cardiovascular Computed Tomography, the American College of Radiology, the American Heart Association, the American Society of Echocardiography, the American Society of Nuclear Cardiology, the North American Society for Cardiovascular Imaging, the Society for Cardiovascular Angiography and Interventions, and the Society for Cardiovascular Magnetic Resonance. Circulation 2010;122:e NakazatoR, Park HB, Berman DS, Gransar H, Koo BK, Erglis A et al. Noninvasivefractional flow reserve derived from computed tomography angiography for coronary lesions of intermediate stenosis severity: results from the DeFACTO study. Circ Cardiovasc Imaging 2013;6: de Graaf MA, Broersen A, Kitslaar PH, Roos CJ, Dijkstra J, Lelieveldt BP et al. Automatic quantification and characterization of coronary atherosclerosis with computed tomography coronary angiography: cross-correlation with intravascular ultrasound virtual histology. Int J Cardiovasc Imaging 2013;29: Nicol ED, Stirrup J, Roughton M, Padley SP, Rubens MB. 64-Channel cardiac computed tomography: intraobserver and interobserver variability (part 1): coronary angiography. J Comput Assist Tomogr 2009;33: Papadopoulou SL, Neefjes LA, Schaap M, Li HL, Capuano E, van dergiessen AGet al. Detection and quantification of coronary atherosclerotic plaque by 64-slice multidetector CT: a systematic head-to-head comparison with intravascular ultrasound. Atherosclerosis 2011;219: Voros S, Rinehart S, Qian Z, Vazquez G, Anderson H, Murrieta L et al. Prospective validation of standardized, 3-dimensional, quantitative coronary computed tomographic plaque measurements using radiofrequency backscatter intravascular ultrasound as reference standard in intermediate coronary arterial lesions: results from the ATLANTA (assessment of tissue characteristics, lesion morphology, and hemodynamics by angiography with fractional flow reserve, intravascular ultrasound and virtualhistology, and noninvasive computed tomography in atherosclerotic plaques) I study. JACC Cardiovasc Interv 2011;4: Fischer C, Hulten E, Belur P, Smith R, Voros S, Villines TC. CoronaryCT angiography versus intravascular ultrasound for estimation of coronary stenosis and atherosclerotic plaque burden: a meta-analysis. J Cardiovasc Comput Tomogr 2013;7: Leber AW, Becker A, Knez A, von Ziegler F, Sirol M, Nikolaou K et al. Accuracy of 64-slice computed tomography to classify and quantify plaque volumes in the proximal coronary system: a comparative study using intravascular ultrasound. J Am Coll Cardiol 2006;47: Brodoefel H, Burgstahler C, Heuschmid M, Reimann A, Khosa F, Kopp A et al. Accuracyof dual-source CT inthecharacterisation of non-calcified plaque: use of acolourcoded analysis compared with virtual histology intravascular ultrasound. Br J Radiol 2009;82: Decramer I, Vanhoenacker PK, Sarno G, VanHoe L, Bladt O, WijnsWet al. Effects of sublingualnitroglycerin oncoronary lumendiameterand numberof visualizedseptal branches on 64-MDCT angiography. Am J Roentgenol 2008;190: Chun EJ, Lee W, Choi YH, Koo BK, Choi SI, Jae HJ et al. Effects of nitroglycerin on the diagnostic accuracy of electrocardiogram-gated coronary computed tomography angiography. J Comput Assist Tomogr 2008;32: Hur J, Kim YJ, Lee HJ, Nam JE, Choe KO, Seo JS et al. Quantification and characterization of obstructive coronary plaques using 64-slice computed tomography: a comparison with intravascular ultrasound. J Comput Assist Tomogr 2009;33: Stein PD, Yaekoub AY, Matta F, Sostman HD. 64-slice CT for diagnosis of coronary artery disease: a systematic review. Am J Med 2008;121: Kruk M, Noll D, Achenbach S, Mintz GS, Pregowski J, Kaczmarska E et al. Impact of coronary artery calcium characteristics on accuracy of CT angiography. JACC Cardiovasc Imaging 2014;7:49 58.