Submillisievert coronary CT angiography with adaptive prospective ECG-triggered sequence acquisition and iterative reconstruction in patients with high heart rate on the dual-source CT

Abstract

PURPOSE: To assess the application value of submillisievert coronary CT angiography (CCTA) in patients with a high heart rate (HR) acquired with adaptive prospective ECG-triggered sequence acquisition and iterative reconstruction on the secondary generation dual-source CT.

MATERIALS AND METHODS: A total of 120 consecutive high-HR patients suspected with coronary artery disease underwent CCTA and invasive coronary angiography (ICA) within two weeks. Patients were randomly assigned into three groups: group A (n = 40), where the patients underwent retrospectively ECG-triggered acquisition CCTA at 100 kVp; group B (n = 40), where the patients received adaptive prospective ECG-triggered sequence acquisition at 100 kVp; and group C (n = 40), where the patients performed adaptive prospective ECG-triggered sequence acquisition at 80 kVp with iterative reconstruction. The mean CT values, signal noise ratios (SNR) and contrast noise ratios (CNR) in the ascending aorta and coronary arteries of the three groups were measured and compared. The image quality and radiation dose among the three groups were compared. The consistency of displaying the coronary stenosis of each group was assessed compared with the results of ICA as the gold standard.

RESULTS: There was no significant difference in gender, age and body mass index (BMI) (all P > 0.05). The mean attenuations, SNRs and CNRs in the ascending aorta and coronary artery were not significantly different between group A and group B (P > 0.05). The mean attenuations of group C were significantly higher than group A and group B (P < 0.01), but the image noise and CNR were significantly lower in group C (P < 0.01). The number of appreciable segments among the three groups was not significantly different on a per-segment and per-vessel basis (P > 0.05). The subjective image quality among the three groups was not significantly different (P > 0.05). With the ICA result as a reference standard, there was good consistency in the evaluation of the coronary stenosis degree between CCTA and ICA (r > 0.75), as well as in the assessment of the coronary stenosis rate using the Bland— Altman analysis. The mean radiation dose in group B was half of that in group A. Moreover, the mean radiation dose in group C was less than one sixth of that in group A and less than 1 mSv (0.7±0.2 mSv).

CONCLUSIONS: For patients with high HR, adaptive prospective ECG-triggered sequence acquisition on the FLASH dual-source CT results in equal image quality and half of the radiation dose reduction compared with retrospectively ECG-triggered spiral acquisition at the same tube voltage (100 kVp) and same R-R interval of exposure. In addition, adaptive prospective ECG-triggered sequence acquisition combined with low tube voltage and iterative reconstruction can further reduce the radiation dose to the submillisievert level without compromising image quality and the accuracy of assessing the coronary stenosis degree, and can be popularized as a routine technique.

Keywords

Dual-source computed tomography coronary angiography adaptive prospective ECG-triggering iterative reconstruction high heart rate radiation dose

1 Introduction

Invasive coronary angiography (ICA) is the relative gold standard for assessing coronary stenosis. It is, however, invasive and carries certain risks which are not easily accepted by patients. With the rapid development of computed tomography (CT) technology, coronary CT angiography (CCTA) has become a reliable and noninvasive imaging technology for coronary artery disease [1 –3]. Retrospectively, electrocardiogram (ECG)-triggered spiral acquisition is usually used for patients with a high heart rate (HR) (>80 beats per minute, bpm) to ensure the image quality and diagnosis accuracy of CCTA. However, the relatively high radiation dose associated with retrospectively ECG-triggered spiral acquisition has raised concerns. The prospective ECG-triggered sequence acquisition on the first dual-source CT can reduce the radiation dose substantially because of the narrow scanning window and it has been gradually used in CCTA examination [4 –6]. However, it is only suitable for patients with low and regular HR. Due to the free control of the scan range of the total dose R-R interval, the improved time resolution of 75 ms and the wide coverage of the 128-slice detector, the adaptive prospective ECG-triggered sequence acquisition on the second dual-source CT (FLASH CT) can be used for CCTA examination of patients with a theoretically high HR. Because some of the patients are not suitable for taking a beta blocker, and other patients who had taken a beta blocker cannot let their heart rate fall below 80 bpm, this study focuses on the application value in patients with a high heart rate using the adaptive prospective ECG-triggered sequence acquisition. In addition, sinogram-affirmed iterative reconstruction (SAFIRE), which has been implanted in an adaptive sequence scan protocol, can drastically reduce image noise and radiation dose of CCTA [7]. Therefore, the aim of our study was to assess the image quality, radiation dose and diagnosis accuracy of CCTA with adaptive prospective ECG-triggered sequence acquisition combined with SAFIRE compared with CCTA with retrospectively ECG-triggered spiral acquisition, and with adaptive prospective ECG-triggered sequence acquisition without SAFIRE to investigate the feasibility and reliability of low dose CCTA for patients with high HR and to broaden the application of low dose CCTA on dual source Flash CT.

2 Materials and methods

2.1 Study subjects

The study was approved by the local ethics board and written informed consent was obtained from all patients. From January 2012 to June 2014, we consecutively collected patients suspected with coronary artery disease that underwent CCTA examination because of acute or chronic chest pain and performed ICA based on clinical manifestations or CCTA results for further diagnosis or treatment within two weeks. No adverse event was recorded during the CCTA and ICA examination. Inclusion criteria were HR >80 bpm, HR variability < 10 bpm and body mass index (BMI) of 20–25 kg/m². Exclusion criteria were cardiac arrhythmia, allergy to iodinated contrast agent, severe impaired renal function and liver function and known coronary cardiac diseases (including myocardial infarction, coronary angioplasty, coronary artery bypass grafts and coronary artery stent implantation). Patients unable to hold their breath were also excluded from this study.

Finally, a total of 120 patients were enrolled in our study. These patients were randomly assigned into three groups according to the random number table: 40 patients underwent CCTA with retrospectively ECG-triggered acquisition at 100 kVp (group A), 40 patients underwent CCTA with adaptive prospective ECG-triggered sequence acquisition at 100 kVp (group B) and 40 patients underwent CCTA with adaptive prospective ECG-triggered sequence acquisition at 80 kVp with iterative reconstruction (group C).

2.2 CT scanning protocol

All examinations were performed on a second-generation dual-source CT system (Definition Flash, Siemens Healthcare, Forchheim, Germany). Patients were strictly instructed to hold their breath by simulating the scanning procedure. Five minutes before the CCTA examination, 0.5 mg of sublingual nitroglycerin spray was administered. After the initial scan, the coronary calcium score scan was performed with a scanning range from the level of the tracheal bifurcation to the diaphragm (mean scanning length of 15 mm). Then, the bolus-tracking contrast enhanced scan was conducted with one circular region-of-interest (ROI) positioned at the root of the ascending aorta. Image acquisition automatically started 6 seconds after the signal attenuation in the ROI reaching the predefined threshold of 100 Hounsfield units (HU). The scanning range was adjusted to be as short as possible to reduce the scanning length along the Z axis according to the images of the coronary calcium score scan. An 18-gauge IV cannula was placed in a superficial vein in the right antecubital fossa. About 70 ml of iopromide (Bayer HealthCare, Pharmaceuticals, Berlin, Germany, 370 mg I/ml) were injected by a dual-syringe injector (Optivantage, Canada) at a flow rate of 5 ml/s followed by 40 ml of saline flush with the same flow rate.

CCTA was performed with retrospectively ECG-triggered acquisition at 100 kVp in group A, adaptive prospective ECG-triggered acquisition at 100 kVp in group B and adaptive prospective ECG-triggered sequence acquisition at 80 kVp in group C respectively (Fig. 1a and 1b). The other scan parameters were kept constant. The tube current was set at 180∼300 mAs and modulated at full dose during 35–45% of the RR interval and at 4% of the full dose for the rest of the RR interval. The pitch varied between 0.33 and 0.44 depending on the heart rate. CCTA data were reconstructed with filtered back projection (FBP) in group A and B and iterative reconstruction in group C (Fig. 1c). All images were reconstructed from 35% to 45% of the R-R interval in 2% increments and the optimal reconstruction phase was used for coronary evaluation.

2.3 Invasive coronary angiography

ICA examinations were performed on the digital flat angiography machine (GE HealthCare, America, Innova3100). The images were acquired with a speed of 25 frames/s and stored in digital subtraction mode. ICA was performed using the conventional Judkins technique. Coronary arteries were imaged by injecting iopromide (Bayer HealthCare, Pharmaceuticals, Berlin, Germany, 370 mg I/ml) at 6–8 ml per injection. The projection position for the left coronary artery was 50° left anterior oblique and 30°Cranial/caudal oblique, 50° right anterior oblique and 30°Cranial/caudal oblique. The projection position for the right coronary artery was 45° left anterior oblique and 30° left anterior oblique.

2.4 Image analysis

The images were reconstructed with a 0.75-mm slice thickness and 0.5-mm increment. FBP series were reconstructed with the B26f kernel and SAFIRE series with the corresponding I26f kernel. All data sets were transferred to a workstation (Multi Modality Workplace, Siemens Medical Solutions, Erlangen, Germany) for post processing. Two independent radiologists with more than 10 years of CCTA experience assessed the curved planar reformations (CPR). In all discordant cases, disagreements were resolved by consensus.

In addition, intravascular attenuations of the ascending aorta and coronary artery were measured with circular ROIs on transverse or multi-planar reformation (CPR) images. The attenuations of the ascending aorta were measured using a manually defined circular ROI measuring 100 mm² at the level of the left main coronary artery (LMA). The attenuations of the proximal right coronary artery (RCA), proximal LMA, proximal left anterior descending artery (LAD) and proximal left circumflex artery (LCX) were also measured on MPR images with a circular ROI measuring 2–5 mm² (a mean of 3 mm²). The attenuation of the coronary artery was defined as the mean of the CT values of the four coronary arteries (RCA, LMA, LAD and LCX). ROIs were drawn as large as possible while avoiding calcifications, plaques and stenoses. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated with the following formulas [8]: $SNR = {CT}_{HU 1} / SD$ $CNR = ({CT}_{HU 1} - {CT}_{HU 2}) / SD$ where CT_HU1 is the mean attenuation of the target artery, CT_HU2 is the mean attenuation of the adjacent perivascular tissue, and SD is the standard deviation of the CT attenuation in the target artery.

All data sets were transferred to Inspace and Circulation software to assess plaques and stenoses in each segment. Coronary arteries were divided into 15 segments for analysis of CCTA data based on the American Heart Association [9]. The RCA was defined as including segments 1–4, the LMA segment 5, the LAD segments 6–10, and the LCX segments 11–15. The intermediate artery was designated as segment 16, if present. The image quality of each segment was evaluated on a 4-point scale as follows [10]: Excellent (grade 1), absence of motion artifacts or minimal noise; Good (grade 2), minor motion artifacts with the performance of local blurring or minimal structural discontinuity of the vessel wall; Adequate (grade 3), moderate motion artifacts with the performance of edge blurring or structural discontinuity of the vessel wall; and Poor (grade 4), marked motion artifact, high image noise, prominent structural discontinuity making evaluation impossible. Images with a score ≤2 were regarded as good and with a score ≤3 as diagnostic.

Curved planar reformat (CPR) views were used to evaluate the coronary artery stenosis rate. All segments with a diameter <1.5 mm were excluded. Vessel diameters were measured on CPRs oriented perpendicularly to the vessel’s centerline. Two senior radiologists calculated the stenosis rate using the following formulas: the stenosis rate (%) = 1–diameter of the narrowest lumen / (the sum of the proximal diameter and distal diameter of the stenosis lumen/2) × 100% [9, 10].

Two senior cardiologists blinded to the results of the CCTA evaluated the coronary stenosis degree on ICA images using the formulas mentioned above. The standard of the coronary stenosis degree for CCTA and ICA was mild degree stenosis, a lumen reduction of less than 50%; moderate degree stenosis, a lumen reduction of 50–75%; severe degree stenosis, a lumen reduction of 75–99%; and occlusion, a lumen reduction of more than 99% [11]. Patients with lumen reduction >50% were considered to have coronary artery disease.

2.5 Radiation dose estimation

Volume CT dose index (CTDIvol) and the dose length product (DLP) for each patient were recorded. The effective dose was calculated by multiplying DLP by a conversion factor for the chest (0.014 mSv×mGy-1×cm–1) according to the European Working Group for Guidelines on Quality Criteria in CT [12]. The radiation dose for each patient was the sum of the radiation dose of the initial scan, coronary calcification score scan and contrast enhanced scan.

2.6 Statistical analysis

Statistical analyses were performed using SPSS software version 17.0 (SPSS Inc. Chicago, IL, USA). The appreciable rate was calculated by dividing the total number of segments by the number of segments scored 1–3. Quantitative variables were expressed as mean±SD. The mean CT value, SNR, CNR, subjective image quality score and radiation dose were compared using the ANOVA test. If a statistical difference was present, inter-group differences were analyzed by using LSD-t text. Kappa analysis was used to assess the inter-observer agreement of the subjective image quality evaluation. The correlation between CCTA and ICA in evaluating coronary artery stenosis was assessed by a Pearson analysis and the consistency was assessed by the Bland-Altman analysis. P values <0.05 were considered to indicate statistically significant differences. An r > 0.75 in the Pearson analysis was regarded as having a good correlation.

3 Results

3.1 Study population

There were no statistically significant differences in gender (male, 55%, 52.5% and 47.5% for group A, B and C respectively, P > 0.05), age (56±11 years, 55±13 years and 54±14 years respectively, P > 0.05) and BMI (23.5±3.1 kg/m², 24.2±3.9 kg/m² and 23.8±4.2 kg/m² respectively, P > 0.05) among the three groups. There were also no statistically significant differences in the heart rate of the patients during the examination among the three groups (91±15, 88±16 and 92±16 bpm respectively, P > 0.05).

3.2 Comparison of subjective image quality and appreciable rate

Table 1 illustrates the CT attenuation, SNR and CNR of the ascending aorta and coronary artery of the three groups. The mean attenuations of the ascending aorta and coronary artery were 461.7±38.3 HU and 442.1±47.2 HU for group A, 458.5±39.9 HU and 443.3±49.7 HU for group B and 615.6±73.2 HU and 575.2±83.0 HU for group C, which showed no significant difference between group A and group B (P = 0.78, 0.81), but a significantly higher attenuation in group C (P < 0.01). The mean SNRs of the ascending aorta and coronary artery were 12.2±2.9 and 16.1±3.5 for group A, 11.9±3.1 and 11.1±3.3 for group B and 8.6±3.7 and 7.9±3.1 for group C and the mean CNRs were 11.3±3.1 and 15.0±3.5 for group A, 15.9±4.4 and 14.7±4.0 for group B and 10.9±3.6 and 9.8±3.5 for group C, which showed no significant difference between group A and group B (P£3/40.05,) but a significantly lower SNR and CNR in group C (P < 0.01) (Fig. 2a, 2b and 2c).

A total of 1698 segments in 120 patients were assessed (Table 2). Among the 1698 segments, 1639 segments were scored 3 or lower, which were considered to be diagnostic. The appreciable rates in group A, B and C were 96.8%, 96.4% and 96.2% respectively. The appreciable rate among the three groups was not significantly different on a per-segment and per-vessel basis (P > 0.05).

3.3 Comparison of objective image quality

Diagnostic image quality (score ≤3) was presented in 38 patients (95%) in group A, 39 patients (97.5%) in group B and 38 patients (95%) in group C, which showed no significant difference among the three groups (P > 0.05) and indicated diagnostic image quality in each group (Fig. 3a, 3b and 3c). The inter-observer agreement was moderate to good for objective image quality of each coronary segment (k = 0.72 for group A; k = 0.47 for group B; k = 0.57 for group C) and moderate for each observer (k = 0.55 for observer 1; k = 0.49 for observer 2) (Table 3).

3.4 Comparison of CCTA with ICA

Approximately 47 patients (14 patients in group A, 17 patients in group B and 16 patients in group C) without coronary artery stenosis were selected by CCTA and ICA. Coronary arteries with different degrees of stenosis indicated by CCTA and/or ICA were presented in 62, 59 and 65 segments per group.

Among the 62 coronary segments with stenosis in group A, 35 segments were verified to have stenosis ≥50% by ICA, which was the same as the results from CCTA. The difference in stenosis rate measured by CCTA and ICA was –6.1±15.6% (95% confidence intervals, CI:–24.6% to 36.8%). There was good consistency in the evaluation of the coronary stenosis degree between CCTA and ICA (n = 62, r = 0.838, 95% CI: 0.75–0.90, P < 0.001) (Fig. 4a). The Bland-Altman diagram (Fig. 4b) for detecting the coronary stenosis degree by CCTA and ICA showed that 4.8% (3/62) of the points were located outside of the boundaries of 95% of the consistency. The maximum absolute difference value between CCTA and ICA in evaluating the stenosis degree of the coronary artery was 35%, which was lower than the mean value of CCTA and ICA (55.3%). Therefore, good consistency was shown in the evaluation of the coronary stenosis degree between CCTA and ICA (Fig. 5a-5f).

Among the 65 coronary segments with stenosis in group C, 36 segments were verified to have stenosis ≥50% by ICA, which was the same as the results from CCTA. The difference in stenosis rate measured by CCTA and ICA was 6.5±14.0% (95% CI:–21.0–33.9%). Good consistency in the evaluation of the coronary stenosis degree between CCTA and ICA (n = 65, r = 0.864, 95% CI: 0.79–0.92, P < 0.001) (Fig. 6a) was shown. The Bland-Altman diagram (Fig. 6b) for detecting the coronary stenosis degree by CCTA and ICA showed that 4.6% (3/65) of the points were located outside of the boundaries of 95% of the consistency. The maximum absolute difference value between CCTA and ICA in evaluating the coronary stenosis degree artery was 31%, which was lower than the mean value of CCTA and ICA (58%). Therefore, good consistency was shown in the evaluation of the coronary stenosis degree between CCTA and ICA (Fig. 7a-7d).

Among the 59 coronary segments with stenosis in group B, 31 segments were verified to have stenosis ≥50% by ICA. The difference in the stenosis rate measured by CCTA and ICA was 7.7±13.5% (95% CI: –18.6–34.1%). There was good consistency in the evaluation of the coronary stenosis degree between CCTA and ICA (n = 59, r = 0.865, 95% CI: 0.78–0.92, P <0.001). The Bland-Altman diagram for detecting the coronary stenosis degree by CTCA and ICA showed good consistency in the evaluation of the coronary stenosis degree between CCTA and ICA.

3.5 Comparison of the radiation dose

Table 4 shows the radiation dose comparison among the three CTA groups. The CTDIvol, DLP and ED were significantly different among the three groups, which showed the highest mean effective dose in group A and the lowest mean effective dose in group C. Compared with group A, the radiation doses of groups B and C were reduced by 50% and 83.3% respectively. Moreover, the mean radiation dose in group C was lower than 1 mSv (0.7±0.2 mSv) and the lowest one was 0.26 mSv.

4 Discussion

Retrospectively, ECG-triggered low-pitch acquisition was the most often used scan protocol for patients with high HR to ensure the image quality and success rate of CCTA. However, more and more attention has been paid on the relatively high radiation dose of this scan protocol. Various approaches have been used to reduce the radiation dose associated with CCTA, including low tube voltage, ECG-dependent tube current modulation, and prospective ECG-triggering [13 –15]. Among these techniques, prospective ECG-triggering is the most effective method to reduce the radiation dose [16]. Hirai et al. showed that 50–80% of radiation dose reduction could be achieved in CCTA with the use of prospective ECG-triggering for patients with low HR and with slightly higher image quality [17]. Prospective ECG-triggered CCTA has been verified to have a high diagnostic accuracy for patients with low HR compared with ICA [16, 18]. However, the conventional prospective ECG-triggered sequence scan protocol (in single source CT with 64 or more slices and the first generation dual-source CT) can only acquire imaging data in the predefined narrow phase of the cardiac cycle (200 ms), which does not allow the reconstruction of the image before and after this phase to make up for the data acquisition loss of a narrow time window, resulting in a low success rate for patients with high HR. Adaptive prospective ECG-triggering sequence acquisition on the second dual-source CT with a temporal resolution of 75 ms can control the acquisition time window and adjust it to the total RR interval, realizing the restructuring phase adjustment. Therefore, the potential application of adaptive prospective ECG-triggering sequence acquisition in patients with high heart rates becomes possible [19, 20].

Adaptive prospective ECG-triggered sequence acquisition can substantially reduce the radiation dose compared with retrospectively ECG-triggered acquisition at the same total dose R-R interval. However, according to the principle of as low as reasonably achievable (ALARA), new approaches have been researched to reduce the radiation dose without compromising image quality and diagnostic accuracy. It is reported that iterative reconstruction has a potential to reduce image noise by more than 35% compared with FBP [21, 22]. Therefore, CCTA using adaptive prospective ECG-triggered sequence acquisition combined with low tube voltage and iterative reconstruction can further reduce the radiation dose on the premise of providing sufficient diagnostic image quality [23 –25].

The radiation dose, image quality and diagnostic accuracy of CCTA in patients with high HR were compared between study groups (adaptive prospective ECG-triggered sequence acquisition combined with low tube voltage and iterative reconstruction) and control groups (retrospective ECG-triggered acquisition and adaptive prospective ECG-triggered sequence acquisition). With the same predefined total dose R-R interval (full dose during 35–45% of the R-R interval and 4% of the full dose for the rest of the R-R interval) in the three groups, the image noise was higher in the study group compared with the control groups. However, because of the higher CT attenuation and the application of iterative reconstruction in the study group, the image quality and diagnostic accuracy for coronary stenosis were not impaired when compared with the control groups [26]. Thus, there was good consistency between CCTA and ICA in the study group, as well as in the two control groups.

The mean radiation dose in group B (2.4±0.5 mSv) was half of that in group C (4.5±0.4 mSv) with the same predefined total dose R-R interval. The main reason for the lower radiation dose in the adaptive prospective ECG-triggered sequence acquisition was the fewer number of cardiac cycles needed to be scanned during data acquisition compared with retrospectively ECG-triggered acquisition (4-5 cardiac cycles vs. 8–10 cardiac cycles) [27, 28]. In addition, the mean radiation dose in group C was half of that in group B and one-sixth of that in group A, which showed the feasibility and reliability of CCTA using adaptive prospective ECG-triggered sequence acquisition combined with 80 kV and iterative reconstruction. Moreover, in patients with HR >80 bpm, the 35–45% of the total dose R-R interval was shortened, resulting in the reduction of the exposure phase and radiation dose without affecting image quality.

There were several limitations in our study. Firstly, the subgroup analysis in patients with HR>80 bpm was not performed. Secondly, patients with high but irregular HR were not enrolled in this study to ensure the success rate of CCTA, which would be further analyzed in the near future.

Conflict of interests

The authors have no financial conflict of interests.

Authors’ contribution

Pei-Hua Tang and Ben-Jun Du contributed equally to this work as the first co-authors.

Footnotes

Acknowledgments

This work is supported by research grants from the National Natural Science Foundation of China (No.81271629, X Fang; No.31271065, Q Gao; No: 81171385, Q Gao; No.30900313, B Du), Youth Fund of National Natural Science Foundation (No.81101043), Jiangsu Province Natural Science Foundation (No.BK2011178), Open project of Key Laboratory of Nuclear Medcine, Ministry of Health (No.KF200906), and Science and Technology Development Fund of Nanjing Medical University (No.2010NJMUZ64).

References

Sabarudin

, Sun

and Ng

K.H.

, Coronary computed tomography angiography with prospective electrocardiography triggering: A systematic review of image quality and radiation dose, Singapore Med J 54(1) (2013), 15–23.

Hamon

, Morello

, Riddell

J.W.

, et al., Coronary arteries: Diagnostic performance of 16-versus 64-section Spiral CT compared with invasive coronary angiography meta-analysis, Radiology 245(3) (2007), 720–731.

FitzGerald

, Bennett

, Carr

, et al., Cardiac CT: A system architecture study, J Xray Sci Technol 24(1) (2016), 43–65.

Husmann

, Valeta

, Gaemperli

, et al., Feasibility of low-dose coronary CT angiography: Fust experience with prospective ECG-gating, Eur Heart J 29(2) (2008), 191–197.

Earls

J.P.

, Berman

E.L.

, Urban

B.A.

, et al., Prospectively gated transverse coronary CT angiography versus retrospectively gated helical technique: Improved image quality and reduced radiation dose, Radiology 246(3) (2008), 742–753.

Hou

, Yue

, Guo

, et al., Prospectively versus retrospectively ECG-gated 256-slice coronary CT angiography: Image quality and radiation dose over expanded heart rates, Int J Cardiovasc Imaging 28(1) (2012), 153–162.

Hara

A.K.

, Paden

R.G.

, Silva

A.C.

, et al., Iterative reconstruction technique for reducing body radiation dose at CT: Feasibility study, AJR 193(3) (2009), 764–771.

Ripsweden

, Brismar

T.B.

, Holm

, et al., Impact on image quality and radiation exposure in coronary CT angiography: 100 kVp versus 120 kVp, Acta Radiol 51(8) (2010), 903–909.

Alkadhi

, Scheffl

, Desbiolles

, et al., Dual-source computed tomography coronary angiography: Influence of obesity,calcium load,and heart rate on diagnostic accuracy, Eur heart J 29(6) (2008), 766–776.

10.

Leschka

, Stolzmann

, Schmid

F.T.

, et al., Low kilovoltage cardiac dual-source CT: Attenuation, noise, and radiation dose, Eur Radiol 18(9) (2008), 1809–1817.

11.

Mollet

N.R.

, Cademartiri

, Van Mieghem

C.A.

, et al., Hight-resolution spiral computed tomography coronary angiography in patients referred for diagnostic conventional coronary angiography, Circulation 112(15) (2005), 2318–2323.

12.

Mayo

J.R.

and Leipsic

J.A.

, Radiation dose in cardiac CT, AJR 192(3) (2009), 646–653.

13.

Hein

, Meyer

, Hadamitzky

, et al., Prospective ECG-triggered sequential scan protocol for coronary dual-source CT angiography: Initial experience, Int J Cardiovasc Imaging 28(3) (2012), 641–650.

14.

Abada

H.T.

, Larchez

, Daoud

, et al., MDCT of the coronary arteries: Feasibility of low-dose CT with ECG-pulsed tube current modulation to reduce radiation dose, AJR 186(6 Suppl 2) (2006), S387–S390.

15.

Shuman

W.P.

, Branch

K.R.

, May

J.M.

, et al., Prospective verse retrospective ECG gating for 64-detector CT of the coronary arteries: Comparison of image quality and patient radiation dose, Radiology 248(2) (2008), 431–437.

16.

Scheffel

, Alkadhi

, Leschka

, et al., Low-dose CT coronary angiography in the step-and-shoot mode: Diagnostic performance, Heart 94(9) (2008), 1132–1137.

17.

Hirai

, Horiguchi

, Fujioka

, et al., Prospective versus retrospective ECG-gated 64-detector coronary CT angiography: Assessment of image quality, stenosis, and radiation dose, Radiology 248(2) (2008), 424–430.

18.

Blankstein

, Shah

, Pale

, et al., Radiation dose and image quality of prospective triggering with dual source cardiac computed tomography, Am J Cardiol 103(8) (2009), 1168–1173.

19.

Pan

C.J.

, Qian

, Wang

, Tang

X.Q.

, et al., Adaptive prospective ECG-triggered sequence coronary angiography in dual-source CT without heart rate control: Image quality and diagnostic performance, Exp Ther Med 5(2) (2013), 636–642.

20.

Zhang

, Pan

H.H.

and Xu

, Evaluating radiation dose of 128-slice dual-source computed tomographic coronary angiography with different electrocardiogram pulsing models, Zhonghua Yi Xue Za Zhi 91(27) (2011), 1923–1925.

21.

Deak

, Grimm

J.M.

, Treitl

, et al., Filtered back projection, adaptive statistical iterative reconstruction, and a model-based iterative reconstruction in abdominal CT: An experimental clinical study, Radiology 266(1) (2013), 197–206.

22.

Lee

S.H.

, Kim

M.J.

, Yoon

C.S.

, et al., Radiation dose reduction with the adaptive statistical iterative reconstruction (ASIR) technique for chest CT in children: An intra-individual comparison, Eur J Radiol 81(9) (2012), 938–943.

23.

Moscariello

, Takx

R.A.

, Schoepf

U.J.

, et al., Coronary CT angiography: Image quality, diagnostic accuracy, and potential for radiation dose reduction using a novel iterative image reconstruction technique comparison with traditional filtered back projection, Eur Radiol 21(10) (2011), 2130–2138.

24.

LaBounty

T.M.

, Leipsic

, Poulter

, et al., Coronary CT angiography of patients with a normal body mass index using 80 kVp versus 100 kVp: A prospective, multicenter, multivendor randomized trial, AJR 197(5) (2011), w860–w867.

25.

, Wang

and Wang

, Edge-oriented dual-dictionary guided enrichment (EDGE) for MRI-CT image reconstruction, J X-ray Sci Technol 24(1) (2016), 161–175.

26.

Dill

, Deetjen

, Ekinci

, et al., Radiation dose exposure in multislice computed tomography of the coronaries in comparison with conventional coronary angiography, In J Cardiol 124(3) (2008), 307–311.

27.

Pflederer

, Jakstat

, Marwan

, et al., Radiation exposure and image quality in staged low dose protocols for coronary dual source CT angiography: A randomized comparison, Eur Radiol 20(5) (2010), 1197–1206.

28.

Fang

X.M.

, Chen

H.W.

, Hu

X.Y.

, et al., Dual-source CT coronary angiography without heart rate or rhythm control in comparison with conventional coronary angiography, Int J Cardiovasc Imaging 26(3) (2010), 323–331.