Non-diagnostic coronary artery calcification and stenosis: a correlation of coronary computed tomography angiography and invasive coronary angiography

Abstract

Background

Heavy coronary artery calcification (CAC) impairs diagnostic accuracy of coronary computed tomography angiography (cCTA) and is considered to be a major limitation.

Purpose

To investigate the effect of non-evaluable CAC seen on cCTA on clinical decision-making by determining the degree of subsequent invasive testing and to assess the relationship between non-evaluable segments containing CAC and significant stenosis as seen in invasive coronary angiography (ICA).

Material and Methods

The study comprised of 356 patients who underwent cCTA and subsequent ICA within 2 months between 2005 and 2009. Clinical reports were reviewed to identify the indications for referral to ICA. In a subset of 68 patients where non-diagnostic CAC on cCTA and significant stenosis on ICA were present in the same segment, we correlated and analyzed the underlying stenosis severity of the lesion on ICA to the cCTA. Lesions with CAC were analyzed in a standardized fashion by application of reading rules.

Results

Non-diagnostic CAC on cCTA prompted ICA in 5.6% of patients. CAC occurred at the site of maximum stenosis in segments with stenosis <50% (95.9% [47/49]), 50–69% (82.4% [28/34]), 70–99% (64.5% [31/48]), and 100% (33.3% [1/3]). At the point of maximum calcium deposit, non-obstructive disease was present in 61.2%. Application of reading rules resulted in a 44% reduction in non-diagnostic cCTA reads.

Conclusion

Severe CAC may prompt further investigation with ICA. There is less CAC with increasing lesion severity at the point of maximum stenosis. Additional application of reading rules improved non-diagnostic cCTA reads.

Keywords

Non-diagnostic coronary artery calcification (CAC)blooming artifacts coronary computed tomography angiography (cCTA)invasive coronary angiography (ICA)coronary artery stenosis

Introduction

Ex vivo studies have shown that the limited spatial resolution of cardiac computed tomography (CT) leads to significant overestimation of the true extent of coronary calcification in CT images (1). This phenomenon causes extensive coronary artery calcification (CAC) to remain a major source of diagnostic uncertainty in the exclusion of significant coronary luminal narrowing resulting in limited positive predictive value and specificity (2 –7). The severity of calcifications has an impact on the agreement between coronary computed tomography angiography (cCTA) and invasive coronary angiography (ICA) in determining obstructive lesions at the individual coronary segment level (8). However, there is a paucity of data in the literature regarding the underlying stenosis severity on ICA at the same segment location where CAC is present with resultant diagnostic ambiguity on cCTA.

We sought to investigate the effect of non-evaluable CAC on clinical decision-making by determining the degree of subsequent invasive testing based on blooming artifacts caused by calcification as seen on cCTA and further invasive interventions such as percutaneous coronary intervention (PCI) or coronary artery bypass graft surgery (CABG). Additionally, in patients with CAC and obstructive lesions on ICA, we sought to explore both the location of calcific deposits and plaque composition causing the obstruction.

Material and Methods

The study was approved by the human research committee of the institutional review board and complied with the Health Insurance Portability and Accountability Act guidelines. The requirement for informed consent was waived for this study. All authors have no relevant financial disclosures.

Patient population selection

Part A: To assess the impact of non-evaluable CAC on subsequent invasive testing as part of a clinical analysis, we identified 517 consecutive patients at our institution between January 2005 and December 2009 who first underwent cCTA and subsequent ICA within 2 months. Patients were excluded if the primary indication for cCTA was not for assessment of suspected coronary artery disease: coronary bypass graft evaluation (n = 98), pulmonary vein mapping (n = 21), cCTA performed for research purposes (n = 17), and right heart catherization without ICA (n = 25). Thus the study population comprised of 356 patients for clinical analysis. All scans were performed as standard of care.

Part B: Re-assessment of cCTA datasets including a detailed plaque analysis using a dedicated image analysis software was performed in the following subset of patients: To investigate the relationship between CAC and coronary stenosis, 85 patients (85 of 356; 23.9%) were identified from review of cCTA and ICA reports where a non-diagnostic CAC on cCTA and significant obstructive lesions ( ≥50% luminal narrowing) on subsequent ICA were present in the same coronary segment. Seventeen patients were excluded due to the following reasons: incomplete CT datasets (n = 3), ICA images unavailable for analysis (n = 3), severe motion or other artifacts at the site of CAC (n = 11). Thus cCTA image analysis with co-registration with ICA was performed in 68 patients (69 of 356) (Fig. 1).

Fig. 1.

Flow chart representing the outline of the study. Out of the 512 patients who received a cCTA and subsequent coronary angiography within 2 months in our department, only clinically indicated cCTA scans for coronary assessment were taken for further analysis. A clinical analysis (Part A) to assess the impact of non-evaluable CAC on subsequent invasive testing, was performed in 312 patients. A detailed cCTA image analysis (Part B) to examine the relationship between non-evaluable CAC and coronary artery stenosis as seen on ICA, was performed in a subset of 68 patients, where non-evaluable CAC and significant stenosis on ICA was present in the same coronary segment. The information if significant coronary artery stenosis and non-evaluable CAC was present in the same segment was taken from clinical reports.

cCTA protocols

Initially between January 2005 and December 2007, scans were performed using 64-multidetector CT scanner (SOMATOM Sensation 64, Siemens AG, Forchheim, Germany). Between January 2008 and December 2009, scans were performed on a first-generation dual-source scanner (SOMATOM Definition 64, Siemens AG) with 2 × 64 × 0.6 mm collimation and gantry rotation time of 330 ms. Contrast-enhanced cCTA was acquired using retrospective ECG-gated helical acquisition (HR > 65 bpm, sinus rhythm) or prospective ECG-triggered axial acquisition (HR <65 bpm, sinus rhythm). Tube potential of either 100 kV (BMI ≤ 25) or 120 kV (BMI > 25) was used, and the tube current was in the range of 128–950 mAs.

For contrast-enhanced cCTA, a 60–80 mL bolus of non-ionic iodinated contrast material (Iopamidol 370 g/cm³, Isovue 370, Bracco Diagnostics, Princeton, NJ, USA), was administered at a flow rate of 5–6 mL/s into an antecubital vein. Metroprolol (oral or i.v.) was given as to achieve a HR < 65 bpm. Sublingual nitroglycerine (0.6 mg) was administered approximately 5 min prior to scanning unless contraindicated. All datasets were reconstructed using conventional filtered back projection.

Invasive coronary angiography

ICA was performed according to standard techniques via a transradial or transfemoral approach. At least two orthogonal views were acquired for all coronary arteries. Significant coronary stenosis was defined as stenosis ≥50%.

Clinical analysis (Part A)

Identification of non-diagnostic coronary segments on cCTA

Clinical cCTA reports of all 356 patients were reviewed to identify the presence and location of non-diagnostic coronary segments. A segment was reported as non-diagnostic if the clinical reader could not exclude a significant stenosis due to blooming artifacts caused by coronary artery calcification as identified in the reports containing the following terms: “could not exclude significant stenosis,” “suspicious for significant stenosis,” “non-evaluable or non-interpretable,” or “likely with or without significant stenosis.”

Indications for referral to ICA and subsequent invasive procedures

ICA reports and medical records of all 356 patients were examined in consensus by one cardiologist (MS) and one cardiovascular imaging fellow (LCE) to identify ICA indications and, if performed, subsequent PCI or CABG.

ICA indications were grouped into five different categories: (i) acute coronary syndrome; (ii) positive or equivocal results of non-invasive imaging/stress test other than cCTA; (iii) severe symptoms not including acute coronary syndrome such as syncope, arrhythmia, dyspnea on exertion and/or shortness of breath, stable angina pectoris; (iv) preoperative assessment of the coronary arteries; and (v) inconclusive cCTA findings due to blooming artifacts caused by CAC.

Non-diagnostic CAC could only be reported as the main indication for ICA if: (i) the patient did not have ACS or other severe symptoms justifying ICA; (ii) the clinical cCTA report did not mention definite significant (≥50%) luminal narrowing; and (iii) results of other additional testing prior to ICA including cardiac biomarkers, echocardiography, electrocardiogram, nuclear stress, exercise treadmill testing, if performed were negative.

CCTA image analysis (Part B)

As described earlier, a detailed plaque analysis was performed in overall 68 patients (Fig. 1). CT datasets were transferred offline and evaluated on a commercially available workstation (Vitrea 2, version 3.8.1; Vital Image, Minnetonka, MN, USA). All datasets were reconstructed using conventional filtered back projection. The following cCTA reading rules were used for interpreting cCTA calcified lesions: (i) use of curved MPR with rotation of vessel long-axis views; (ii) use of the thinnest slice thickness; (iii) use of a wide window setting (W2000, L200); and (iv) in case of presence of any luminal contrast visible adjacent to calcium complete coronary artery occlusion was excluded.

Non-diagnostic CAC was defined as a totally calcified plaque or predominantly calcified plaque (plaque with less than 50% non-calcified component), in which significant luminal narrowing (>50%) could not be entirely excluded by the cCTA reader solely due to a potential blooming artifact (no signs of motion, streak, or windmill artifacts, incomplete coverage, or increase background noise levels).

An automatic vessel detection tool on curved multi-planar reconstructions and cross-sectional images perpendicular to the vessel was used to initially segment the coronary artery, as described previously (9). Two cardiologists blinded to the clinical and ICA data (GL and WT, with at least 5 years of experience in cardiovascular CT imaging and ICA) first evaluated the cCTA images to determine if the segments containing calcium were interpretable and graded the stenosis severity at the site of calcification. Subsequently, the readers were unblinded to the ICA cine images on a DICOM viewer (Osirix 3.6.1, Geneva, Switzerland) where co-registration between cCTA and ICA was performed using anatomical landmarks such as vessel branch points.

At the coronary segments on cCTA where there was blooming caused by CAC, we determined the stenosis severity on ICA images after co-registration the cCTA and ICA images. Grading of the stenosis severity on ICA images were performed by an experienced interventional cardiologist (GL) divided into the following categories: <50%, 50–69%, 70–99%, and 100%. Furthermore, we identified the location of maximal coronary stenosis reported on ICA and analyzed the corresponding location on cCTA images using the same co-registration technique to determine if: (i) the stenosis on ICA was at the site of CAC on cCTA; (ii) if so, the plaque composition; and (iii) the stenosis severity on cCTA.

Statistical analysis

Continuous variables were expressed as mean ± standard deviations for or median with interquartile range (IQR; 25th and 75th percentiles) where appropriate. Categorical variables were expressed as frequencies or percentages. Statistical calculations were performed using PASW (Predictive Analytics SoftWare) Statistics Version 18.0 (SPSS Inc., Chicago, IL, USA).

Results

Clinical analysis (Part A)

The study population (n = 356) underwent cCTA and subsequent ICA within a mean time frame of 17.8 ± 16.7 days. The mean Agatston score was 438.3 ± 575.4. Fig. 2 outlines the number of subsequent PCI and CABG performed in patients with and without non-diagnostic segments due to CAC on cCTA. Of the 356 patients, 52% (185/356) had ≥1 non-diagnostic segment due to extensive calcification.

Fig. 2.

Clinical work flow chart and subsequent ICA findings in patients with (grey) versus without (white) non-diagnostic segments containing CAC as seen on cCTA. Overall rates of performed PCI and CABG were compared between both groups.

Non-diagnostic CAC as a trigger for ICA and subsequent PCIs and CABGs

Indications for referral to ICA are displayed in Fig. 3. Non-diagnostic segments due to blooming CAC on cCTA prompted ICA in 5.6 % of patients (20 out of 356).

Fig. 3.

Indications for referral to ICA. In our study cohort of 356 patients who underwent non-invasive and subsequent invasive diagnostic testing, we identified indications for undergoing ICA. Non-diagnostic CAC was the main reason for referral to ICA in 5.6% (20/356). However, in the majority of these cases (3.1%, i.e. 11/356), the segment containing non-diagnostic CAC did not have an underlying stenosis.

The overall rate of PCI and CABG was 35.7% (127/356) and 10.7% (38/356), respectively. PCI was performed in 43.3% (80/185) of patients with at least one non-diagnostic segment due to extensive CAC on cCTA, compared to 27.5% (47/171) in patients with no non-diagnostic segments (Fig. 2). Review of cCTA and ICA reports revealed that 27.6% (n = 35) of PCI were performed in segments with non-diagnostic CAC while 72.4% of PCI (n = 92) were performed in segments that were not hampered by CAC.

Of all patients where non-diagnostic CAC prompted further evaluation using ICA (n = 20; 5.6%), five patients (1.4%) were free of significant obstructive CAD (<50% luminal narrowing) while 15 patients (4.2%) had at least one significant coronary stenosis (≥50% luminal narrowing). Non-diagnostic segments with extensive CAC did not have an underlying stenosis on ICA in 11 out of 20 patients (i.e. 3.1% of overall cohort) while in nine out of 20 patients, non-diagnostic CAC was located in the same segment as the coronary obstruction (i.e. 2.5% of overall cohort) (Fig. 3).

CCTA image analysis (Part B)

CAC and coronary stenosis

Baseline characteristics of the subgroup of 68 patients used for image analysis are shown in Table 1. The proportion of lesions with CAC causing stenosis in a segment-based and lesion-based analysis are displayed in Table 2.

Table 1.

Cohort (n = 68) for detailed image analysis: baseline characteristics.

Baseline characteristics
Patients (n)	68
Age (years)	66.8 ± 11.5
Male gender	72.1% (49/68)
Ethnicity
– White	94.1% (64/68)
– Asian	2.9% (2/68)
– Black	0% (0/68)
– Other	2.9% (2/68)
Weight (lbs)	192.4 ± 43.6
BMI (kg/m²)	29.8 ± 5.8
Angina symptoms
– Typical angina	54.4% (37/68)
– Atypical angina	27.9% (19/68)
Heart rate (bpm)	63.3 ± 8.6
Betablockade	64.7% (44/68)
Hyperlipidemia	83.8% (57/68)
History of smoking	36.7% (25/68)
Hypertension	86.8% (59/68)
Diabetes mellitus	32.4% (22/68)
Family history of CAD	42.6% (29/68)

Table 2.

Percentage of non-diagnostic calcified lesions causing stenosis in a segment- versus a lesion-based analysis. In the lesion-based analysis, non-diagnostic CAC as seen on cCTA was taken as the reference point and stenosis at the point of maximum calcium deposit (non-diagnostic site) was determined qualitatively based on ICA.

Degree of stenosis on ICA	Lesions with CAC (%)
Degree of stenosis on ICA	Segment-based analysis	Lesion-based analysis
<50% (n = 49)	36.6% (49/134)	61.2% (82/134)
50–69% (n = 34)	25.4% (34/134)	17.2% (23/134)
70–99% (n = 48)	35.8% (48/134)	20.1% (27/134)
Total occlusions (n = 3)	2.2% (3/134)	1.5% (2/134)

Location of CAC and plaque composition at maximum stenosis

The location of the maximum calcium deposit of CAC with respect to stenosis on cross-sectional and longitudinal CT images are shown in Table 3. On longitudinal views, CAC occurred at the site of maximum stenosis in segments with stenosis <50% (95.9% [47/49]), 50–69% (82.4% [28/34]), and 70–99% (64.5% [31/48]). In total occlusions, the location of CAC with respect to the occlusion site was equally distributed proximal and distal to as well as at the site of occlusion.

Table 3.

Location of CAC with respect to maximum stenosis on a cross-sectional and longitudinal view. Stenosis as assessed on ICA was taken as the reference point and location of any CAC (not necessarily the site of maximum calcium deposit) in relation of maximum coronary obstruction was determined.

Location of CAC on CCTA	Degree of stenosis on ICA
Location of CAC on CCTA	<50%	50–69%	70–99%	100%
Longitudinal view
Upstream (within 5 mm proximal)	4.1% (2/49)	5.9% (2/34)	18.8% (9/48)	33.3% (1/3)
At the site of max. stenosis	95.9% (47/49)	82.4% (28/34)	64.5% (31/48)	33.3% (1/3)
Downstream (within 5 mm distal)	0% (0/49)	11.8% (4/34)	14.6% (7/48)	33.3% (1/3)
Cross-sectional view
Lumen	28.5% (14/49)	52.9% (18/34)	56.3% (27/48)	66.6% (2/3)
Vessel wall	71.4% (35/49)	47.1% (16/34)	41.7% (20/48)	33.3% (1/3)

On cross-sectional views, 71.4% (35/49) of non-obstructive lesions (stenosis <50%), CAC was located closer to the vessel wall rather than the lumen (28.5% [14/49]). In obstructive lesions, CAC was most often seen adjacent to the lumen in the following frequencies: 50–69% stenosis: 52.9% (18/34); 70–99% stenosis: 56.3% (27/48); 100% stenosis/total occlusion: 66.6% (2/3).

Plaque composition at the site of maximum stenosis is shown in Fig. 4a.

Fig. 4.

(a) Plaque composition by cCTA at the site of maximum coronary artery stenosis. (b) Composition of partially calcified plaques on cCTA, the site of maximum coronary artery stenosis.

Fig. 4b shows the ratio of predominantly calcified to non-calcified plaque in partially calcified plaques at the highest grade of luminal narrowing. We found that as the stenosis severity category increases, there is a pattern where the calcific component decreases.

Assessment of non-diagnostic CAC after application of reading rules

Analysis of the clinical cCTA reports found 134 lesions which were reported to be non-diagnostic due to calcification. However, after reassessment of corresponding cCTA datasets using the four reading rules mentioned in the “Methods” section, we found that only 75/134 lesions were considered non-diagnostic.

Discussion

The purpose of this study on cCTA was to investigate the impact of non-diagnostic coronary artery calcification (CAC) on clinical decision-making by determining the degree of subsequent invasive testing and to assess the relationship between non-evaluable segments containing CAC and significant stenosis as seen in ICA. It has been shown that with the appropriate patient population, cCTA can serve as a “gate keeper” for ICA and is associated with reduced downstream testing potentially saving unnecessary radiation exposure (10,11). We found that in a real-world population, patients with non-diagnostic calcified segments on cCTA had more PCIs performed compared to patients with diagnostic CAC segments. In the proportion of patients who underwent ICA due solely to non-diagnostic segments caused by extensive CAC on cCTA, just under half of these patients (9 out of 356 patients, e.g. 2.5%) had an underlying significant coronary stenosis on ICA at the site of the non-diagnostic segments (see black proportion in Fig. 3). Furthermore, we found that the largest calcific deposit may not be in the same location as the lesion responsible for the obstruction. As the stenosis severity on ICA increases, the calcific component of partially calcified coronary plaque seen on CT decreases. This study also presents four cCTA reading rules which after application reduced the number of non-diagnostic cases due to CAC.

With respect to the clinical outcome, the presence of heavy calcification is regarded as a sign of stability as calcified plaques are less prone to rupture and biomechanically stable (12), while the presence of non-calcified plaque is associated with acute coronary syndrome (13). Autopsy studies demonstrated that the calcification in thin-cap fibroatheroma and plaque rupture tends to be speckled (<2 mm) or fragmented (>2 mm, <5 mm) whereas extensively calcified plaques are less prone to a change in atheroma volume (14,15). Additionally, Mauriello et al. found that calcification was not identified as an independent determinant of unstable plaques in multivariate analysis (16).

In a first site by site comparison of coronary calcification between cardiac CT (electron beam CT: EBCT) and ICA published by Kitajima et al., the authors concluded that extensive calcification was more frequently associated with luminal narrowing than spotty calcification (17). On the other hand, it has been proposed that the majority of calcified plaques undergo remodeling and do not lead to luminal narrowing (13,18). Our study supports the latter observation with the majority of PCI performed in segments not hampered by CAC and the majority of segments with non-diagnostic CAC did not cause significant luminal narrowing. As shown previously (18), we also found that the presence of calcified lesions decreased with increasing grade of luminal narrowing and in partially calcified plaque, the calcific component decreased as the stenosis severity increased. Of note, the largest calcific deposit may not necessarily be the culprit responsible for causing significant coronary obstruction. In 61.2% of non-diagnostic calcified or mixed lesions, the maximum calcium deposit (i.e. non-diagnostic site) did not have an underlying obstructive lesion and significant stenosis was more often found at a different area of the segment, suggesting that severe stenosis occurs at the border between calcified and non-calcified plaque components or completely at the non-calcified plaque component (Fig. 5).

Fig. 5.

Three different retrospective gated dual-source CT studies showing presence of mixed-plaque in the mid LAD. Dense calcified portion (black line) is estimated to cause significant stenosis (a–c). Corresponding invasive coronary angiography images show that calcification (black line) caused non-significant stenosis. Coronary obstruction was caused by non-calcified portion (white arrows) of mixed-plaques (d–f).

Technological developments in the current scanner generation like iterative reconstruction techniques have shown to significantly decrease image noise (19). As a result, lower image noise enables the use of sharper edge-enhancement kernels, which may contribute to an improved signal-to-noise and contrast-to-noise ratio in the presence of extensive coronary artery calcification (19,20). Despite lacking iterative reconstruction techniques, our findings suggest that an improvement of the diagnostic ability of cCTA in the presence of heavy CAC might be possible by using four specific CT reading rules by which we were able to achieve a 44% reduction in non-diagnostic cCTA reads.

This retrospective study is limited by an observational design at a single center. The threshold for calling plaques non-interpretable due to high calcium burden may be different compared to other centers. Furthermore, co-registration between cCTA and ICA was only carried out in a subset of patients. Stenosis severity was assessed in categories rather than absolute degrees of stenosis for both ICA and CCTA. However, if a reliable and robust CT quantitative coronary analysis (QCA) tool was available, it may potentially yield more accurate results when compared to QCA on ICA cine-angiograms. In addition, we acknowledge that iterative reconstruction techniques were not available at our institution during the study period. This may account for an overall higher rate of non-evaluable segments.

In conclusion, severe CAC remains a challenge in the interpretation of cCTA which may prompt further investigation with ICA. On a pathophysiological level, there is less CAC with increasing lesion severity at the point of maximum stenosis. Application of strict reading rules in the presence of extensive CAC in calcified or partly calcified plaques improves non-diagnostic cCTA reads and may potentially reduce additional invasive testing. Larger multi-centered studies are needed to further validate these findings.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Sarwar

Rieber

Mooyaart

, et al. Calcified plaque: measurement of area at thin-section flat-panel CT and 64-section multidetector CT and comparison with histopathologic findings. Radiology 2008; 249: 301–306.

Alkadhi

Scheffel

Desbiolles

, et al. Dual-source computed tomography coronary angiography: influence of obesity, calcium load, and heart rate on diagnostic accuracy. Eur Heart J 2008; 29: 766–776.

Bamberg

Abbara

Schlett

, et al. Predictors of image quality of coronary computed tomography in the acute care setting of patients with chest pain. Eur J Radiol 2010; 74: 182–188.

Brodoefel

Burgstahler

Tsiflikas

, et al. Dual-source CT: effect of heart rate, heart rate variability, and calcification on image quality and diagnostic accuracy. Radiology 2008; 247: 346–355.

Hoffmann

Nagurney

Moselewski

, et al. Coronary multidetector computed tomography in the assessment of patients with acute chest pain. Circulation 2006; 114: 2251–2260.

Vanhecke

Madder

Weber

, et al. Development and validation of a predictive screening tool for uninterpretable coronary CT angiography results. Circ Cardiovasc Imaging 2011; 4: 490–497.

Zhang

Levin

Halpern

, et al. Accuracy of MDCT in assessing the degree of stenosis caused by calcified coronary artery plaques. Am J Roentgenol 2008; 191: 1676–1683.

Vavere

Arbab-Zadeh

Rochitte

, et al. Coronary artery stenoses: accuracy of 64-detector row CT angiography in segments with mild, moderate, or severe calcification–a subanalysis of the CORE-64 trial. Radiology 2011; 261: 100–108.

Kang

Park

, et al. CT comparison of visual and computerised quantification of coronary stenosis according to plaque composition. Eur Radiol 2011; 21: 712–721.

10.

Meijs

Meijboom

Prokop

, et al. Is there a role for CT coronary angiography in patients with symptomatic angina? Effect of coronary calcium score on identification of stenosis. Int J Cardiovasc Imaging 2009; 25: 847–854.

11.

Cheezum

Hulten

Taylor

, et al. Cardiac CT angiography compared with myocardial perfusion stress testing on downstream resource utilization. J Cardiovasc Comput Tomogr 2011; 5: 101–109.

12.

Huang

Virmani

Younis

, et al. The impact of calcification on the biomechanical stability of atherosclerotic plaques. Circulation 2001; 103: 1051–1056.

13.

Motoyama

Sarai

Harigaya

, et al. Computed tomographic angiography characteristics of atherosclerotic plaques subsequently resulting in acute coronary syndrome. J Am Coll Cardiol 2009; 54: 49–57.

14.

Otsuka

Sakakura

Yahagi

, et al.

Has our understanding of calcification in human coronary atherosclerosis progressed?

Arterioscler Thromb Vasc Biol 2014; 34: 724–736.

15.

Virmani

Joner

Sakakura

. Recent highlights of ATVB: calcification. Arterioscler Thromb Vasc Biol 2014; 34: 1329–1332.

16.

Mauriello

Servadei

Zoccai

, et al. Coronary calcification identifies the vulnerable patient rather than the vulnerable Plaque. Atherosclerosis 2013; 229: 124–129.

17.

Kajinami

Seki

Takekoshi

, et al. Coronary calcification and coronary atherosclerosis: site by site comparative morphologic study of electron beam computed tomography and coronary angiography. J Am Coll Cardiol 1997; 29: 1549–1556.

18.

Park

Jung

Choi

, et al. Coronary CT angiography in patients with high calcium score: evaluation of plaque characteristics and diagnostic accuracy. Int J Cardiovasc Imaging 2011; 27(Suppl. 1): 43–51.

19.

Hell

Bittner

Schuhbaeck

, et al. Prospectively ECG-triggered high-pitch coronary angiography with third-generation dual-source CT at 70 kVp tube voltage: feasibility, image quality, radiation dose, and effect of iterative reconstruction. J Cardiovasc Comput Tomogr 2014; 8: 418–425.

20.

Takahashi

Kimura

Umezawa

, et al. Comparison of adaptive statistical iterative and filtered back projection reconstruction techniques in quantifying coronary calcium. J Cardiovasc Comput Tomogr 2015; 10: 61–66.