Clinical Relevance of Coronary Computed Tomography Angiography Beyond Coronary Artery Stenosis

Key words

cardiac - CT-angiography - angiography

Introduction

In 1979, Godfrey Hounsfield and Allan Cormack were awarded the Nobel Prize in Physiology and Medicine for the development of computer-assisted tomography. In his Nobel Prize acceptance speech, Godfrey Hounsfield said “A further promising field may be the detection of the coronary arteries” [1]. In the years that followed, the challenge of capturing the beating heart has driven innovation in the field at a remarkable pace. As life expectancy throughout the world has risen, the global burden of cardiovascular disease has followed suit [2]. It is therefore fitting that coronary computed tomography angiography (CCTA) should advance to face it.

Despite many advances in the field, clinicians often focus on the ability of CCTA to predict the severity of coronary artery stenosis, perhaps oblivious to its full potential. The totality of CCTA’s capabilities is too numerous to adequately discuss in a single review. Computational fluid dynamic algorithms enable a functional assessment of stenotic lesions, with the potential to reduce unnecessary invasive angiography [3]. Mapping changes in perivascular fat attenuation has the potential to enhance cardiac risk prediction and change how we target the prescribing of preventative therapies [4]. In this review, we will focus on the ways in which CCTA can evaluate atherosclerotic plaque to provide clinically relevant information, over and above simply answering the question “how narrow is that blood vessel?”

Basic visual assessment of plaque

Early angiographic studies demonstrated that culprit lesions in myocardial infarction were not always associated with severe stenosis [5]. Visual assessment of the coronary arteries, therefore, requires more than just an assessment of the severity of stenosis. Atherosclerosis describes the process of plaque deposition which is an immune-mediated inflammatory process, exacerbated by metabolic risk factors [6]. Plaques likely to rupture are characterized by inflamed and attenuated fibrous caps covering large and necrotic lipid cores [7] [8]. As such, along with describing the distribution of disease, a central goal of basic visual assessment is to determine the composition of the coronary plaque, as this contributes to plaque vulnerability.

On CCTA, atherosclerotic plaques can be classified as calcified, non-calcified, or mixed plaques. While calcified plaque has always been easy to detect, detecting non-calcified plaque accurately has only been possible in the era of multi-slice CT scanning [9] [10]. Kopp et al. were the first to demonstrate the ability of CT to noninvasively characterize lesion morphology and composition [11]. Multiple studies have since demonstrated the excellent correlation between CT attenuation density and plaque characterization as determined by intravascular ultrasound [12] [13]. Culprit lesions in acute coronary syndromes are more likely to be non-calcified than calcified [14]. There are also interesting sex-based differences in plaque type with women having more non-calcified plaques and men having more calcified plaques [15] [16]. However, the prognostic implications of basic plaque classification are less certain. In 458 patients presenting with acute chest pain but without an acute coronary syndrome, Nance et al. found that the occurrence of MACE (major adverse cardiovascular events) at 13 months was higher in those with mixed plaques rather than non-calcified or calcified plaques [17]. However, in 1584 stable patients undergoing CCTA for suspected coronary artery disease, Hadamitzky et al. found that plaque classification as non-calcified, calcified, or mixed did not improve risk stratification over assessment of stenosis severity [18]. One of the important limiting factors of this form of plaque assessment is that observer variability for visual plaque analysis has been shown to be poor [19]. Moreover, this basic assessment does not allow for quantification of the extent of disease.

Semiquantitative assessment of plaque burden

A distinct advantage of CCTA over other noninvasive imaging modalities is the ability to derive a measure of atherosclerotic burden throughout the coronary tree. This is particularly important as Maddox et al. found that those with multi-vessel nonobstructive disease have a similar prognosis to those with single vessel obstructive disease [20]. A variety of semi-quantitative scores have been proposed which aim to summarize the results of CCTA into a single metric that communicates plaque burden ([Table 1]) [21].

Table 1 Comparison of semiquantitative score progression.
Tab. 1 Vergleich des semiquantitativen Score-Verlaufs.

Score	Explanation	Sample case: patient with one 70 % lesion in the left main stem
Coronary Artery Disease – Reporting and Data System (CADS-RADS)	score 0–5 depending on the severity of the worst stenosis: 0–0 %, no CAD 1–1–24 %, minimal non-obstructive 2–25–49 %, mild non-obstructive 3–50–69 %, moderate stenosis 4–70–99 %, severe stenosis A – > 50 % LMS B – 3-vessel ≥ 70 % 5–100 %, total coronary occlusion	CAD-RADS Score: 4
Segment Involved Score (SIS)	score depending on the absolute number of segments with any disease based on the 17-segment coronary tree model. 0 – no coronary artery disease 1 – coronary plaque present continuous score, range: 0–16	SIS Score: 1
Segment Stenosis Score (SSS)	score depending on severity of stenosis in each segment based on the 17-segment coronary tree model. 0 – no coronary artery disease 1 – mild plaque 2 – moderate plaque 3 – severe plaque continuous score, range: 0–48	SSS Score: 3
CT-adapted Leaman Score (CT-LeSc)	weighted score based on: location of coronary plaque 5.0–6.0 for LMS depending on dominance 1.0–3.5 for LAD segments and branches depending on dominance 0.5–2.5 for LCx segments depending on dominance 0.5–1.0 for RCA segments depending on dominance severity of stenosis multiply score by 1 for obstructive multiply score by 0.615 for non-obstructive coronary plaque composition multiple score by 1.5 for non-calcified or mixed plaque multiple score by 1.0 for calcified plaque	CT-LeSc Score if right dominant: 5 if calcified 7.5 if non-calcified CT-LeSc Score of left dominant: 6 if calcified 9 if non-calcified

CAD- coronary artery disease, LMS- left main stem, LAD- left anterior descending artery, LCx- left circumflex artery, RCA- right coronary artery.

The “Segment Involved Score” (SIS) is a semi-quantitative measure of the extent of coronary artery disease throughout the coronary tree. In the SIS, segments are scored 0 or 1 based on the presence or absence of plaque, irrespective of the degree of stenosis. Meta-analyses have established extent of disease as determined by the SIS is a strong and independent predictor of cardiovascular mortality [22]. Moreover, recent results from the CONFIRM (Coronary CT Angiography Evaluation for Clinical Outcomes: an International Multicenter) registry suggest that an SIS > 5 provides more prognostic information for MACE than traditional cardiovascular risk factors such as hypertension or diabetes [23]. Despite this, the SIS is limited due to the lack of consideration given to stenosis severity. Recently, more comprehensive scores such as the CT-adapted Leaman score have been shown to improve prognostic stratification. In accounting for lesion locale, plaque composition, and degree of stenosis, the CT-adapted Leaman score performs considerably better than SIS [24]. However, these scores do not account for more advanced visual assessment of high-risk plaque and only provide an estimate of plaque burden rather than true quantification.

Visual assessment of high-risk plaque

While the extent of disease and severity of stenosis are undoubtedly important, they do not provide any information on the vulnerability of a plaque to rupture. Autopsy studies have established the thin cap fibroatheroma as the histological precursor of plaque rupture [25]. CCTA correlates of the thin-cap fibroatheroma include positive remodeling (a positive change in vessel diameter at the plaque site compared to a normal-appearing proximal segment), low attenuation plaque (< 30 Hounsfield units), spotty calcification (calcification < 3 mm in size) and the “napkin-ring sign” ([Fig. 1]). These have been established by correlating CCTA findings to intravascular ultrasound (IVUS) and optical coherence tomography (OCT) findings [26] [27]. Early work by Motoyama et al. demonstrated the significant association of positive remodeling, low attenuation plaque, and spotty calcification with plaque rupture events in patients who had suffered an acute coronary syndrome [28]. Moreover, in their follow-up study, they demonstrated that positively remodeled segments with low-attenuation plaque were significantly more likely to result in acute coronary syndromes [29] [30]. As an individual plaque feature, the “napkin-ring sign” correlates with histological findings of central necrotic lipid cores surrounded by fibrous tissue [31] and demonstrated excellent specificity in identifying advanced lesions [32].

Fig. 1 CT coronary angiogram of diseased coronary artery, lumen highlighted in blue. A Normal proximal left main stem measuring 5.1 mm × 5.4 mm. B Calcified lesion distal to the left main stem, positively remodeled, measuring 6.5 mm × 5.8 mm. C High-risk plaque (napkin-ring sign) in the proximal left anterior descending artery, vessel measuring 6.2 mm × 5.7 mm.

Abb. 1 CT-Koronarangiogramm einer erkrankten Koronararterie, Lumen blau hervorgehoben. A Normaler proximaler linker Hauptstamm mit den Abmessungen 5,1 mm × 5,4 mm. B Verkalkte Läsion distal des linken Hauptstamms, positiv remodelliert, mit Abmessungen von 6,5 mm × 5,8 mm. C Hoch-Risiko-Plaque (Serviettenring-Zeichen) in der proximalen linken anterioren absteigenden Arterie, Gefäßgröße 6,2 mm × 5,7 mm.

Several studies have subsequently built on these findings ([Table 2]). The ICONIC (Incident COroNary Syndromes Identified by Computed tomography) case-control sub-study of the CONFIRM registry found that high-risk plaque features predict future acute coronary syndromes independent of, and better than, atherosclerotic plaque burden and the number of obstructed vessels [33]. The ROMICAT-2 (Rule Out Myocardial Infarction using Computer-Assisted Tomography) trial found in troponin- and electrocardiogram-negative patients presenting with chest pain to the emergency department, the presence of high-risk plaque on CCTA increased the likelihood of myocardial infarction independent of clinical risk assessment and extent of coronary disease [34]. In the PROMISE (Prospective Multicenter Imaging Study for Evaluation of Chest Pain) trial, of the 4415 patients with stable chest pain who underwent CCTA, 15 % had high-risk plaques [35]. Patients with high-risk plaques had an increased risk of MACE (hazard ratio 2.73, 95 % confidence interval, 1.89 to 3.93), which was independent of cardiovascular risk score and the presence of significant stenosis. Interestingly, the presence of high-risk plaque was a more important predictor of events in women and younger patients. In a prospective cohort study of 1469 patients, Feuchtner et al. found that the strongest predictors of cardiovascular events over an 8-year period were low-attenuation plaque and the napkin-ring sign [36]. Together these studies show that high-risk plaque features provide important prognostic information, over and above traditional assessments, and they are now part of the Society of Cardiovascular Computed Tomography CAD-RADS reporting guidelines [37].

Table 2 Key studies assessing quantitative and qualitative plaque on CCTA.
Tab. 2 Wichtige Studien zur Bewertung der quantitativen und qualitativen Plaque mit CCTA.

Visual assessment of high-risk plaques		Quantitative assessment of plaque
Author, date	Findings	Author, date	Findings
Motoyama et al., 2007 [29]	HRP is an independent predictor of acute coronary syndrome	ROMICAT, 2012 [43]	ACS patients have a higher volume of plaque with low CT density (< 90 HU)
ROMICAT-2, 2014 [33]	HRP in troponin-negative and ECG-indeterminate patients with chest pain increased the likelihood of MI independent of clinical risk assessment and atherosclerotic plaque burden	ROMICAT II, 2015 [44]	ACS patients have a higher volume of plaque with a low CT density (< 30HU and < 60 HU)
Motoyama et al., 2015 [56]	HRP is an independent predictor of acute coronary syndrome at 4 years. Plaque progression by serial CCTA is an independent predictor of acute coronary syndrome	Nadjiri et al., 2016 [47]	non-calcified plaque volume and low-attenuation plaque volume are predictive of MACE at 5 years
Feuchtner et al., 2016 [35]	High-risk low-attenuation plaque and the napkin-ring sign are the most powerful predictors of MACE over long-term follow-up (8 years)	ICONIC, 2018 [32]	cross-sectional plaque burden, fibro-fatty and necrotic core volumes were higher in ACS patients than controls. All three were significant predictors of ACS
PROMISE, 2018 [34]	HRP is associated with an increased risk of MACE after adjustment for cardiovascular risk and presence of significant stenoses	PARADIGM, 2018 [49]	progression of atherosclerosis is slowed by statin therapy. Females are more responsive to statin compared to men
ICONIC, 2018 [32]	HRP predicted future ACS independent of and better than number of obstructive vessels and atherosclerotic plaque burden	de Knegt, 2019 [45]	ACS patients have a higher plaque volume, with more fibro-fatty plaque and less densely calcified plaque
SCOT-HEART, 2019 [37]	HRP is associated with a worse prognosis but not independent of coronary calcium score	SCOT-HEART, 2020 [48]	low-attenuation plaque burden is the strongest predictor of fatal or non-fatal myocardial infarction

ACS, acute coronary syndrome; ECG, electrocardiogram; HRP, high-risk plaque; MACE, major adverse cardiovascular event; MI, myocardial infarction.

However, the inter-observer variability for the identification of high-risk plaque features has been shown to be only fair, which limits their use in clinical practice [19]. In the Scottish Computed Tomography of the HEART (SCOT-HEART) trial, patients with positive remodeling or visually assessed low-attenuation plaque had a three-fold increase in the rate of fatal or non-fatal myocardial infarction (hazard ratio 3.01, 95 % confidence interval 1.61 to 5.63, p = 0.001).[38] However, this was not independent of the coronary artery calcium score, a surrogate marker of the overall plaque burden. It is likely that these high-risk plaque features are of particular importance early on after imaging, but they continue to evolve and potentially stabilize with time. In addition, these high-risk plaques are common on CCTA, occurring in 15 % to 50 % of patients depending on their presenting symptoms [33] [35] [38]. Thus, the presence of visually assessed high-risk plaque can be used to identify patients at an increased risk of myocardial infarction, but not all patients with high-risk plaque will undergo myocardial infarction.

Quantitative plaque assessment

While semi-quantitative scores provide important prognostic information, they remain a surrogate for actual measurements of plaque volume and burden. With advances in computing technology, we are now able to quantitatively assess plaque subtypes on CCTA based on their attenuation density. Total plaque volume can be measured as well as plaque subtypes including calcified and noncalcified (fibrous, fibrofatty, and necrotic or low-attenuation) plaque. This technology could be used to identify patients with an increased plaque burden or an increased burden of particular high-risk plaque subtypes ([Fig. 2]). In addition, assessing the progression of plaque subtypes in such detail can facilitate our understanding of the impact of medications on the atherosclerotic process.

Fig. 2 Quantitative plaque assessment (Autoplaque, Los Angeles, US) of a stenotic mid left anterior descending artery. Blue represents the lumen, calcified plaque volume (highlighted in yellow) 3.0 mm³, non-calcified plaque volume (highlighted in red) 154.3 mm³, low attenuation plaque volume (highlighted in orange) 8.0 mm³.

Abb. 2 Quantitative Plaque-Bewertung (Autoplaque, Los Angeles, US) einer stenotischen mittleren linken anterioren absteigenden Arterie. Blau repräsentiert das Lumen, verkalktes Plaquevolumen (gelb hervorgehoben) 3,0 mm³, nicht verkalktes Plaque-Volumen (rot hervorgehoben) 154,3 mm³, gering verkalktes Plaque-Volumen (orange hervorgehoben) 8,0 mm³.

Early iterations of plaque quantification software were time-consuming, manual processes that could not differentiate between low-attenuation and non-calcified plaques [39]. Accordingly, semi-automated software has now been developed that significantly reduces the time needed to quantify plaque burden. The software demonstrates improved repeatability and reproducibility over manual quantification, especially in patients with low to intermediate disease burden [40] [41]. Moreover, improved algorithms allow for more a precise description of low-attenuation plaque burden which correlated better with intravascular ultrasound [42] [43]. These advances have streamlined our ability to measure plaque burden and progression on serial imaging.

In patients with acute chest pain, the ROMICAT [44] and ROMICAT II [45] studies found that patients with acute coronary syndromes had a larger volume of plaque with a low attenuation density. De Knegt et al. showed that compared to asymptomatic patients and patients with acute chest pain without acute coronary syndrome, patients with acute coronary syndromes had a higher total plaque volume and volume of fibrofatty and necrotic core plaque, but a lower volume of densely calcified plaque [46]. These studies highlight the differences in plaque subtypes found in patients with different clinical presentations.

Several studies have established the particular importance of low-attenuation plaque, associated with the necrotic core of the thin-cap fibroatheroma. In the ICONIC sub-study of the CONFIRM trial, increased cross-sectional plaque burden, fibrofatty plaque volume, and necrotic core volume were all associated with increased risk of subsequent acute coronary syndrome in 234 patients with acute coronary syndrome compared to matched control pairs [33]. Interestingly, they found that there were no sex-based differences in calcified plaque volume, but women had lower fibrous and fibrofatty plaque volume compared to men [47]. Nadjiri et al. found that in 1168 patients undergoing CCTA for suspected coronary artery disease the volume of non-calcified plaque and low-attenuation plaque was higher in patients that experienced MACE during 5 years of follow-up [48].

Recently, a post hoc analysis of the SCOT-HEART trial showed the primacy of low-attenuation plaque burden in the prediction of future fatal or non-fatal myocardial infarction [49]. The total plaque burden and the burden of all sub-types of plaque were higher in patients who suffered subsequent myocardial infarction after 4.7 years of follow-up. Low-attenuation plaque burden was the strongest predictor of subsequent myocardial infarction (adjusted hazard ratio per doubling 1.60, 95 % confidence interval 1.10 to 2.34, p = 0.014), over and above the cardiovascular risk score, coronary artery calcium score, and coronary artery stenoses. Patients with a low-attenuation plaque burden above 4 % were at a particularly high risk for subsequent myocardial infarction (hazard ratio 4.65, 95 % confidence interval 2.06 to 10.5, p < 0.001). Thus, in patients presenting with stable chest pain, quantitative plaque burden provides better prognostic information than classic markers of cardiovascular risk.

The PARADIGM (Progression of AtheRosclerotic PlAque DetermIned by Computed TomoGraphic Angiography) trial was a large, prospective, observational study that evaluated temporal changes in plaque characteristics utilizing semi-automated plaque quantification software [50]. This trial demonstrated that statins not only resulted in slower rates of progression of non-calcified plaque volume, but also reduced the risk of positive remodeling and high-risk plaque formation. Importantly, they were able to quantitatively assess the impact of statins on the whole coronary tree. Progression of subclinical atherosclerosis was slowed in vessels beyond the proximal segments that are usually assessed by intravascular ultrasound [50]. The authors were also able to describe sex-based differences in plaque composition (high-risk plaque was more common in men than women) and plaque progression (female sex was associated with greater progression of calcified plaque and reduced progression of non-calcified plaque) [51].

Together these studies show the power of quantitative plaque assessment in identifying patients at risk for subsequent cardiac events and the impact of medications on plaque progression. However, as the quantification of plaque is based on CT attenuation density, it is limited by scan image quality, including motion artifacts, image noise, and stair-step artifacts. Its use in advanced, calcific disease and the primary prevention populations have yet to be evaluated. Although plaque quantification does not require new hardware or special scanning techniques, dedicated software is required, and individual readers require training to be able to adjust vessel and plaque contours if required. At present, quantitative plaque analysis remains a valuable research technique. Current research studies have shown that quantitative plaque assessment can be used as an endpoint in drug trials. Future research studies will assess whether management based on quantitative plaque analysis can improve outcomes.

Functional assessment of coronary stenosis – Computed Tomography Fractional Flow Reserve

A common criticism of coronary assessments made by CCTA is the lack of associated functional information. When this is combined with the tendency to overestimate the severity of calcified lesions, patients may require more “downstream” noninvasive imaging to clarify the significance of a particular lesion on CCTA [52]. Fractional flow reserve (FFR) measures the change in pressure across a coronary lesion under pharmacological stress using specially designed pressure-sensitive wires. Multiple randomized controlled trials have demonstrated that when such physiological measurements are used to guide coronary revascularization, outcomes are improved and unnecessary coronary interventions are reduced [53] [54].

CTFFR (computed tomography fractional flow reserve) utilizes computational fluid dynamics and models physiological conditions of hyperemia to produce an estimate of the invasive FFR. The diagnostic accuracy of these calculations was directly confirmed in several trials, where CTFFR results were compared directly to invasive FFR [55] [56]. Correlation between both were good, and moreover the diagnostic accuracy of CTFFR was significantly better than with CCTA alone for the identification of hemodynamically significant lesions. The PLATFORM study (Prospective LongitudinAl Trial of FFRct: Outcome and Resource Impacts) went on to demonstrate a 61 % reduction in non-obstructive coronary arteries on invasive angiography and a significant reduction in costs compared to usual care with equivalent clinical outcomes [3]. The advantages of CTFFR over other noninvasive tests are clear, in that it provides anatomical and functional information, without the requirement to perform additional imaging or radiation exposure. However, at present its use remains limited due to the need for careful selection on the basis that image quality can greatly affect the reliability of results [57].

Future developments – Radiomics & machine learning

Fundamentally, radiological images are large 3-dimensional vaults of data, with each voxel representing unique tissue-dependent measurements. As we image structures with higher resolution, these datasets have grown exponentially in size, providing us with ever increasing quantities of information. Radiomics aims to extract further information from these datasets by using mathematical techniques to extract higher dimension data such as spatial interrelationships and textural information. Machine learning, a branch of artificial intelligence, can be used to mine these datasets to identify radiomic patterns associated with an increased risk of cardiac events. Kolossvary et al. showed that radiomic features can identify high-risk plaques with good diagnostic accuracy compared to IVUS and 18F-sodium fluoride PET, better than visual assessment alone [58].

There are numerous applications of both supervised and unsupervised machine learning in CCTA, including the identification and quantification of atherosclerotic plaque. The identification of calcified plaque on CT using deep learning has been widely studied, particularly on non-contrast images, but the automatic identification of non-calcified and high-risk plaque subtypes is more challenging [59]. Recently, a deep learning algorithm that identified CCTA without calcification has been proposed as a method to help prioritize work lists [60]. Further advancements in machine learning to automate plaque analysis will reduce the time to perform this analysis and increase its application in clinical practice.

Machine learning techniques can also be used to analyze the complex interactions between multiple parameters in large datasets. For example, when machine learning was used to combined clinical and CCTA data from the CONFIRM registry, it performed significantly better than clinical risk scores (Framingham) and CCTA severity scores (SIS and SSS) at predicting all-cause mortality (p < 0.001 for all) [61]. In another example, machine learning was used to integrate quantitative CCTA plaque metrics including plaque measurements, diameter stenosis, and contrast density difference (maximal difference in luminal attenuation per unit area) and was shown to be better at predicting ischemia by FFR over any individual measure [62].

The potential applications of machine learning include precision diagnostics, automated risk stratification and enhanced health economy, thus reducing healthcare costs by saving clinicians valuable time. At present, the clinical applications are limited, but machine learning is an exciting avenue for future research and is likely to become an integral part of clinical practice over the coming decades.

Conclusion

The technological advances both in how we acquire and how we interpret CCTA images have undergone rapid and sustained innovation, particularly in the last decade. However, there is a considerable lag in routine clinical practice for CCTA which remains largely based on lumenogram. While stenosis severity is one important variable, this review highlights the critical importance of quantifying and classifying coronary artery plaque to substantially improve the diagnostic and prognostic potential of CCTA.

• References

• 1 Hounsfield G. Computed Medical Imaging. Science 1980; 210: 22-28
• 2 Murray CJL, Lopez AD. Global mortality, disability, and the contribution of risk factors: Global Burden of Disease Study. The Lancet 1997; 349: 1436-1442
• 3 Douglas PS, De Bruyne B, Pontone G. et al. 1-Year Outcomes of FFRCT-Guided Care in Patients With Suspected Coronary Disease: The PLATFORM Study. J Am Coll Cardiol 2016; 68: 435-445
• 4 Oikonomou EK, Marwan M, Desai MY. et al. Non-invasive detection of coronary inflammation using computed tomography and prediction of residual cardiovascular risk (the CRISP CT study): a post-hoc analysis of prospective outcome data. The Lancet 2018; 392: 929-939
• 5 Hackett D, Davies G, Maseri A. Pre-existing coronary stenosis in patients with first myocardial infarction are not necessarily severe. Eur Heart J 1988; 9: 1317-1323
• 6 Hansson GK. Mechanisms of disease: Inflammation, Atherosclerosis and Coronary Artery disease. N Engl J Med 2005; 352: 1685-1695
• 7 van der Wal AC, Becker A, van der Loos CM. et al. Site of Intimal Rupture or Erosion of Thrombosed Coronary Atherosclerotic Plaques Is Characterized by an Inflammatory Process Irrespective of the Dominant Plaque Morphology. Circulation 1994; 89: 36-44
• 8 Davies MJ, Thomas A. Thrombosis and Acute Coronary Artery Lesions in Sudden Cardiac Ischaemic Death. N Engl J Med 1984; 310: 1137-1140
• 9 Hurlock GS, Higashino H, Mochizuki T. History of cardiac computed tomography: single to 320-detector row multislice computed tomography. Int J Cardiovasc Imaging 2009; 25 (Suppl. 01) 31-42
• 10 Hausleiter J, Meyer T, Hadamitzky M. et al. Prevalence of noncalcified coronary plaques by 64-slice computed tomography in patients with an intermediate risk for significant coronary artery disease. J Am Coll Cardiol 2006; 48: 312-318
• 11 Kopp AF, Schroeder S, Baumbach A. et al. Non-invasive characterisation of coronary lesion morphology and composition by multislice CT: first results in comparison with intracoronary ultrasound. Eur Radiol 2001; 11: 1607-1611
• 12 Motoyama S, Kondo T, Anno H. et al. Atherosclerotic Plaque Characterization by 0.5-mm- Slice Multislice Computed Tomographic Imaging Comparison With Intravascular Ultrasound. Circulation Journal 2007; 71: 363-366
• 13 Leber AW, Becker A, Knez A. 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: 672-677
• 14 Hoffmann U, Moselewski F, Nieman K. et al. Noninvasive assessment of plaque morphology and composition in culprit and stable lesions in acute coronary syndrome and stable lesions in stable angina by multidetector computed tomography. J Am Coll Cardiol 2006; 47: 1655-1662
• 15 Blaha MJ, Nasir K, Rivera JJ. et al. Gender differences in coronary plaque composition by coronary computed tomography angiography. Coronary Artery Disease 2009; 20: 506-512
• 16 Plank F, Beyer C, Friedrich G. et al. Sex differences in coronary artery plaque composition detected by coronary computed tomography: quantitative and qualitative analysis. Neth Heart J 2019; 27: 272-280
• 17 Nance JW, Schlett CL, Schoepf UJ. et al. Incremental Prognostic Value of Different components of coronary atherosclerotic Plaque at cardiac CT angiography beyond coronary calcification in Patients with acute chest Pain. Radiology 2012; 264: 679-690
• 18 Hadamitzky M, Taubert S, Deseive S. et al. Prognostic value of coronary computed tomography angiography during 5 years of follow-up in patients with suspected coronary artery disease. Eur Heart J 2013; 34: 3277-3285
• 19 Maroules CD, Hamilton-Craig C, Branch K. et al. Coronary artery disease reporting and data system (CAD-RADS(TM)): Inter-observer agreement for assessment categories and modifiers. J Cardiovasc Comput Tomogr 2018; 12: 125-130
• 20 Maddox TM, Stanislawski MA, Grunwald GK. et al. Nonobstructive coronary artery disease and risk of myocardial infarction. JAMA 2014; 312: 1754-1763
• 21 van den Hoogen IJ, van Rosendael AR, Lin FY. et al. Coronary atherosclerosis scoring with semiquantitative CCTA risk scores for prediction of major adverse cardiac events: Propensity score-based analysis of diabetic and non-diabetic patients. J Cardiovasc Comput Tomogr 2020; 14: 251-257
• 22 Ayoub C, Erthal F, Abdelsalam MA. et al. Prognostic value of segment involvement score compared to other measures of coronary atherosclerosis by computed tomography: A systematic review and meta-analysis. J Cardiovasc Comput Tomogr 2017; 11: 258-267
• 23 van Rosendael AR, Bax AM, Smit JM. et al. Clinical risk factors and atherosclerotic plaque extent to define risk for major events in patients without obstructive coronary artery disease: the long-term coronary computed tomography angiography CONFIRM registry. Eur Heart J Cardiovasc Imaging 2020; 21: 479-488
• 24 Andreini D, Pontone G, Mushtaq S. et al. Long-term prognostic impact of CT-Leaman score in patients with non-obstructive CAD: Results from the COronary CT Angiography EvaluatioN For Clinical Outcomes InteRnational Multicenter (CONFIRM) study. Int J Cardiol 2017; 231: 18-25
• 25 Virmani R, Burke AP, Kolodgie FD. et al. Vulnerable Plaque: The Pathology of Unstable Coronary Lesions. Journal of Interventional Cardiology 2002; 15: 439-446
• 26 Kashiwagi M, Tanaka A, Kitabata H. et al. Feasibility of noninvasive assessment of thin-cap fibroatheroma by multidetector computed tomography. JACC Cardiovasc Imaging 2009; 2: 1412-1419
• 27 Obaid DR, Calvert PA, Brown A. et al. Coronary CT angiography features of ruptured and high-risk atherosclerotic plaques: Correlation with intra-vascular ultrasound. J Cardiovasc Comput Tomogr 2017; 11: 455-461
• 28 Motoyama S, Kondo T, Sarai M. et al. Multislice computed tomographic characteristics of coronary lesions in acute coronary syndromes. J Am Coll Cardiol 2007; 50: 319-326
• 29 Motoyama S, Sarai M, Harigaya H. et al. Computed tomographic angiography characteristics of atherosclerotic plaques subsequently resulting in acute coronary syndrome. J Am Coll Cardiol 2009; 54: 49-57
• 30 Motoyama S, Ito H, Sarai M. et al. Plaque Characterization by Coronary Computed Tomography Angiography and the Likelihood of Acute Coronary Events in Mid-Term Follow-Up. J Am Coll Cardiol 2015; 66: 337-346
• 31 Maurovich-Horvat P, Hoffmann U, Vorpahl M. et al. The napkin-ring sign: CT signature of high-risk coronary plaques?. JACC Cardiovasc Imaging 2010; 3: 440-444
• 32 Maurovich-Horvat P, Schlett CL, Alkadhi H. et al. The napkin-ring sign indicates advanced atherosclerotic lesions in coronary CT angiography. JACC Cardiovasc Imaging 2012; 5: 1243-1252
• 33 Chang HJ, Lin FY, Lee SE. et al. Coronary Atherosclerotic Precursors of Acute Coronary Syndromes. J Am Coll Cardiol 2018; 71: 2511-2522
• 34 Puchner SB, Liu T, Mayrhofer T. et al. High-risk plaque detected on coronary CT angiography predicts acute coronary syndromes independent of significant stenosis in acute chest pain: results from the ROMICAT-II trial. J Am Coll Cardiol 2014; 64: 684-692
• 35 Ferencik M, Mayrhofer T, Bittner DO. et al. Use of High-Risk Coronary Atherosclerotic Plaque Detection for Risk Stratification of Patients With Stable Chest Pain: A Secondary Analysis of the PROMISE Randomized Clinical Trial. JAMA Cardiology 2018; 3: 144-152
• 36 Feuchtner G, Kerber J, Burghard P. et al. The high-risk criteria low-attenuation plaque < 60 HU and the napkin-ring sign are the most powerful predictors of MACE: a long-term follow-up study. Eur Heart J Cardiovasc Imaging 2017; 18: 772-779
• 37 Cury RC, Abbara S, Achenbach S. et al. CAD-RADS(TM) Coronary Artery Disease - Reporting and Data System. An expert consensus document of the Society of Cardiovascular Computed Tomography (SCCT), the American College of Radiology (ACR) and the North American Society for Cardiovascular Imaging (NASCI). Endorsed by the American College of Cardiology. J Cardiovasc Comput Tomogr 2016; 10: 269-281
• 38 Williams MC, Moss AJ, Dweck M. et al. Coronary Artery Plaque Characteristics Associated With Adverse Outcomes in the SCOT-HEART Study. J Am Coll Cardiol 2019; 73: 291-301
• 39 Cheng VY, Nakazato R, Dey D. et al. Reproducibility of coronary artery plaque volume and composition quantification by 64-detector row coronary computed tomographic angiography: an intraobserver, interobserver, and interscan variability study. J Cardiovasc Comput Tomogr 2009; 3: 312-320
• 40 Schuhbaeck A, Dey D, Otaki Y. et al. Interscan reproducibility of quantitative coronary plaque volume and composition from CT coronary angiography using an automated method. Eur Radiol 2014; 24: 2300-2308
• 41 Ovrehus KA, Schuhbaeck A, Marwan M. et al. Reproducibility of semi-automatic coronary plaque quantification in coronary CT angiography with sub-mSv radiation dose. J Cardiovasc Comput Tomogr 2016; 10: 114-120
• 42 Boogers MJ, Broersen A, van Velzen JE. et al. Automated quantification of coronary plaque with computed tomography: comparison with intravascular ultrasound using a dedicated registration algorithm for fusion-based quantification. Eur Heart J 2012; 33: 1007-1016
• 43 Matsumoto H, Watanabe S, Kyo E. et al. Improved Evaluation of Lipid-Rich Plaque at Coronary CT Angiography: Head-to-Head Comparison with Intravascular US. Radiol Cardiothorac Imaging 2019; 1: e190069
• 44 Ferencik M, Schlett CL, Ghoshhajra BB. et al. A computed tomography-based coronary lesion score to predict acute coronary syndrome among patients with acute chest pain and significant coronary stenosis on coronary computed tomographic angiogram. Am J Cardiol 2012; 110: 183-189
• 45 Ferencik M, Mayrhofer T, Puchner SB. et al. Computed tomography-based high-risk coronary plaque score to predict acute coronary syndrome among patients with acute chest pain--Results from the ROMICAT II trial. J Cardiovasc Comput Tomogr 2015; 9: 538-545
• 46 de Knegt MC, Linde JJ, Fuchs A. et al. Relationship between patient presentation and morphology of coronary atherosclerosis by quantitative multidetector computed tomography. Eur Heart J Cardiovasc Imaging 2019; 20: 1221-1230
• 47 Conte E, Dwivedi A, Mushtaq S. et al. Age- and sex-related features of atherosclerosis from coronary computed tomography angiography in patients prior to acute coronary syndrome: results from the ICONIC study. Eur Heart J Cardiovasc Imaging 2021; 22: 24-33
• 48 Nadjiri J, Hausleiter J, Jahnichen C. et al. Incremental prognostic value of quantitative plaque assessment in coronary CT angiography during 5 years of follow up. J Cardiovasc Comput Tomogr 2016; 10: 97-104
• 49 Williams MC, Kwiecinski J, Doris M. et al. Low-Attenuation Noncalcified Plaque on Coronary Computed Tomography Angiography Predicts Myocardial Infarction: Results From the Multicenter SCOT-HEART Trial (Scottish Computed Tomography of the HEART). Circulation 2020; 141: 1452-1462
• 50 Lee SE, Chang HJ, Sung JM. et al. Effects of Statins on Coronary Atherosclerotic Plaques: The PARADIGM Study. JACC Cardiovasc Imaging 2018; 11: 1475-1484
• 51 Lee SE, Sung JM, Andreini D. et al. Sex Differences in Compositional Plaque Volume Progression in Patients With Coronary Artery Disease. JACC Cardiovasc Imaging 2020; 13: 2386-2396
• 52 Morris JR, Bellolio MF, Sangaralingham LR. et al. Comparative Trends and Downstream Outcomes of Coronary Computed Tomography Angiography and Cardiac Stress Testing in Emergency Department Patients With Chest Pain: An Administrative Claims Analysis. Acad Emerg Med 2016; 23: 1022-1030
• 53 Pijls NH, van Schaardenburgh P, Manoharan G. et al. Percutaneous coronary intervention of functionally nonsignificant stenosis: 5-year follow-up of the DEFER Study. J Am Coll Cardiol 2007; 49: 2105-2111
• 54 van Nunen LX, Zimmermann FM, Tonino PAL. et al. Fractional flow reserve versus angiography for guidance of PCI in patients with multivessel coronary artery disease (FAME): 5-year follow-up of a randomised controlled trial. The Lancet 2015; 386: 1853-1860
• 55 Koo BK, Erglis A, Doh JH. et al. Diagnosis of ischemia-causing coronary stenoses by noninvasive fractional flow reserve computed from coronary computed tomographic angiograms. Results from the prospective multicenter DISCOVER-FLOW (Diagnosis of Ischemia-Causing Stenoses Obtained Via Noninvasive Fractional Flow Reserve) study. J Am Coll Cardiol 2011; 58: 1989-1997
• 56 Norgaard BL, Leipsic J, Gaur S. et al. Diagnostic performance of noninvasive fractional flow reserve derived from coronary computed tomography angiography in suspected coronary artery disease: the NXT trial (Analysis of Coronary Blood Flow Using CT Angiography: Next Steps). J Am Coll Cardiol 2014; 63: 1145-1155
• 57 Pontone G, Weir-McCall JR, Baggiano A. et al. Determinants of Rejection Rate for Coronary CT Angiography Fractional Flow Reserve Analysis. Radiology 2019; 292: 597-605
• 58 Kolossvary M, Park J, Bang JI. et al. Identification of invasive and radionuclide imaging markers of coronary plaque vulnerability using radiomic analysis of coronary computed tomography angiography. Eur Heart J Cardiovasc Imaging 2019; 20: 1250-1258
• 59 van Velzen SGM, Lessmann N, Velthuis BK. et al. Deep Learning for Automatic Calcium Scoring in CT: Validation Using Multiple Cardiac CT and Chest CT Protocols. Radiology 2020; 295: 66-79
• 60 van den Oever LB, Cornelissen L, Vonder M. et al. Deep learning for automated exclusion of cardiac CT examinations negative for coronary artery calcium. Eur J Radiol 2020; 129: 109114
• 61 Motwani M, Dey D, Berman DS. et al. Machine learning for prediction of all-cause mortality in patients with suspected coronary artery disease: a 5-year multicentre prospective registry analysis. Eur Heart J 2017; 38: 500-507
• 62 Dey D, Gaur S, Ovrehus KA. et al. Integrated prediction of lesion-specific ischaemia from quantitative coronary CT angiography using machine learning: a multicentre study. Eur Radiol 2018; 28: 2655-2664