Study Of Low Dose Prospectively Triggered Sequential Dual Source CT Coronary Angiography

Background and Objection:Coronary heart disease is one of the most major and fatal diseases that seriously jeopardize human health at home and abroad. It has brought a huge negative social and economic impact. Multi-slice spiral computed tomography coronary angiography (MSCTCA) is a rapid, non-invasive, and accurate tool for the diagnosis and treatment evaluation of coronary heart disease. It has been widely used in daily clinical practice. In spite of its important clinical value, coronary CT angiography (CCTA) may potentially bring some radiation harm to the paitents due to the usage of X-ray during the course of CT scanning. According to literature reports, CCTA using conventional retrospectively triggered scan mode needs a high radiation dosage of 5-30 mSv. Even though no direct evidence confirms the exact relationship between X-ray radiation of CT scan and malignancies, the theory of radiation protection indicates that there is still some potential risk for the health of the body after even a tiny dose of ionizing radiation. The clinical practice of CCTA examination should strictly abide by the world-widely accepted ALARA principle of radiation protection. It is to say that the radiation dosage of CCTA should be as low as reasonably achievable.Image data of CCTA can be acquired with retrospectively electrocardiography-(ECG) triggered spiral scan mode, prospectively ECG-triggered high-pitch spiral scan mode, or prospectively ECG-triggered sequential scan mode. Each mode has its significant impact on radiation dosage of CCTA to a certain extent. As the most conventional scan mode, retrospectively ECG-triggered CCTA needs a high radiation dosage of 5-30 mSv due to the usage of continuous exposure of X-ray during several cardiac cycles. Prospectively triggered high-pitch spiral scan mode is unique to the second and third generation of dual source CT. This scan mode was not equipped in the first generation of dual source CT. The scanning of a whole heart can be completed in just one cardiac cycle with a low radiation dose of less than 1 mSv using prospectively triggered high-pitch spiral scan mode. But there are some limitations for the usage of this high-pitch scan mode. It has a strict requirement of low and stable heart rates. Patients with high or irregular heart rates are not suitable for this scan mode. Prospectively ECG-triggered sequential scan mode, also called "step-and-shoot" scan mode, uses axial protocol to complete the acquisition of image data in several cardiac cylces intermittently. It was convinced that prospectively ECG-triggered sequential scan mode can significantly reduce nearly 70% of radiation dose without the sacrifice of image quality and diagnositic accuracy compared with retrospectively ECG-triggered spiral CCTA. According to several researches, prospectively triggered sequential CCTA using the second generation 2 × 128 slices dual source CT is feasible for patients with high heart rates, irregular heart rates, and even atrial fibrillation compared with retrospectively triggered spiral CCTA. Prospectively triggered sequential scan mode can manually select and automatically adjust the exposure windows according to the condition of patients’heart rates during CCTA scanning. Remarkable reduction of more than 50% of radiation dosage could be achieved while keeping consistent level of image quality and diagnostic ability using prospectively triggered sequential CCTA.The first aim of this paper is to investigate image quality and radiation dose of single-verse multi-phase acquisition protocol for prospectively triggered sequential coronary angiography using the first generation 2 x 64 slices dual source CT. The second aim of this paper is to explore the feasibility, image quality, and radiation dose of prospectively triggered sequential CCTA protocol with an ultra-low tube potential of 70 kV compared with 80 kV protocol using the second generation of 2× 128 slices dual source CT.Methods and materialsStudy 1:Prospectively triggered adaptive sequential CCTA using the first generation 2 x 64 slices dual source CT was performed on 144 patients with low (≤ 70 bpm) and stable (heart rate variability<10 bpm) heart rates. All the patients were randomly arranged in single-phase group (group A) or multi-phase group (group B). A narrow acquisition window of 70%-70% R-R interval was used in group A and a wide acquition window of 65%-75% R-R interval was used in group B. A low tube potential of 100 kV was used for patients with a body weight of<90 kg or body mass index of<30 kg/m2 and a regular tube potential of 120 kV was used for patients with a body weight of≥90 kg or body mass index of≥30 kg/m2. The amount of Iohexol injection (Omnipaque 350,350 mg I/mL; GE Healthcare) was applied based on the body weight (1 mL/kg) of individual patient, followed by 40 mL of 20% blended contrast with saline.All the CCTA images were automatically reconstructed using filtered back projection algorithm with a medium-soft convolution kernel (B26f) after the completion of data acquisition. One set of heart images was acquired at 70% of the R-R interval in group A. In group B, multi-phase data sets, including the data set at 70% R-R interval that was used as a virtual single-phase group (group C), were reconstructed for the selection of the optimal data set of the best phase. The actual acquisition time of each scan block was recorded for the calculation of the acquisition time variability of scan blocks.Two skilled radiologists blindly evaluated the image quality independently on per-segment basis. A 16-segment model proposed by the American Heart Association was used to assess image quality with a 4-point scale (Fig.2):1=excellent; 2=good; 3=fair, diagnosable image quality; 4=poor, non-diagnostic image quality. The objective measurement of image quality was performed independently by another experienced radiologist. A circular region of interests (ROI) were placed in the root of ascending aorta (AO), the pericardial mediastinal fat, the proximal segment of right coronary artery (RCA), left main artery (LM), left anterior descending artery (LAD), and left circumflex artery (LCX). The CT value of each ROI was recorded as image signal intensity. The standard deviation of the ROI in the root of AO was recorded as image signal noise. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated. The volume computed tomography dose index (CTDIvol) and dose length product (DLP) of each patient was recorded to calculate the estimated effective dose (ED) and the size-specific dose estimates (SSDE).Study 2:Prospectively triggered adaptive sequential CCTA using the second generation 2 x 128 slices dual source CT was performed on 100 patients with normal body shape (body mass index< 26 kg/m2) suspected of coronary artery disease. All the patients were randomly arranged in 70 kV group (group A) or 80 kV group (group B). An ultra-low tube potential of 70 kV was used in group A and a low tube potential of 80 kV was used in group B. A minimal acquisition window of 70%-70% R-R interval was used for patients with a heart rate of≤ 70 bpm (heart rate variability< 10 bpm or only having accidental premature beat). A wide acquisition window of 30%-80% R-R interval was used for patients with a heart rate of≤ 70 bpm (heart rate variability≥10 bpm but except accidental premature beat), or for patients with a heart rate between 71 bpm and 80 bpm. A narrow acquisition window of 30%-50% R-R interval was used for patients with a heart rate of>80 bpm. The amount of Iohexol injection (Omnipaque 350,350 mg I/mL; GE Healthcare) was 45 mL, followed by 45 mL saline.All the CCTA images of the best phase were automatically reconstructed using an iterative reconstruction algorithm (SAFIRE) with a medium-smooth convolution kernel (I26f) and a medium strength of 3 after the completion of data acquisition. If the automatically selected best phase showed blurred or stair-step artifacts, data sets at multiple phases were reconstructed with an interval of 5% R-R interval for the manual selection of the best phase. The selection of the optimal data set was performed independently by a senior radiologist at the preprocessing stage.Two experienced radiologists blindly evaluated the image quality independently on per-segment basis. A 16-segment model proposed by the American Heart Association was used to assess image quality with a 4-point scale:1=excellent; 2= good; 3=fair, diagnosable image quality; 4=poor, non-diagnostic image quality. The objective measurement of image quality was performed independently by another experienced radiologist. A circular region of interests were placed in the root of ascending aorta, the pericardial mediastinal fat, the proximal segment of right coronary artery, left main artery, left anterior descending artery, and left circumflex artery. The CT attenuation of each ROI was recorded as image signal intensity. The standard deviation of the ROI in the root of AO was also recorded as image signal noise. The signal-to-noise ratio and contrast-to-noise ratio were calculated. The volume computed tomography dose index and dose length product of each patient was recorded to calculate the estimated effective dose and the size-specific dose estimates.Statistical AnalysisShapiro-Wilk test wasused to test whether the data was normally distributed. A P value of< 0.05 was considered significant.Study 1:The differences in patient demographic characteristics, image acquisition parameters, and estimated radiation dose were compared between group A and group B by using the Student t test or Mann-Whitney U test for continuous variables, and Pearson Chi-square test or Fisher’s exact test for categorical variables. The correlation between heart rate variability and acquisition time variability was tested by Pearson correlation test. Interobserver agreement in subjective image quality assessment was evaluated by using Kappa analysis. Mean image quality grade was compared between group A and group B, and between group A and group C by using Mann-Whitney U test. Mean image quality grade was compared between group B and group C by using Wilcoxon signed-rank test. The objective measurement parameters between group A and group B, and between group A and group C were tested by using Student t test or Mann-Whitney U test. The objective measurement parameters between group B and group C were tested by using paired t test or Wilcoxon signed rank test. The diagnostic rates were compared between group A and group B, and between group A and group C by using Pearson Chi-square test. The diagnostic rates were also compared between group B and group C by using McNemar test.Study 2:The differences in patient demographic characteristics, image acquisition parameters, and estimated radiation dose were compared between group A and group B by using the Student t test or Mann-Whitney U test for continuous variables, and Pearson Chi-square test or Fisher’s exact test for categorical variables. Interobserver agreement in subjective image quality assessment was evaluated by using Kappa analysis. Mean image quality grade was compared between the two groups using Mann-Whitney U test. The diagnostic rates were compared between the two groups by using Pearson Chi-square test. The objective measurement parameters between the two groups were tested by using Student t test or Mann-Whitney U test.Results:Study 1:There were no significant differences in the ratio of male and the usage of 100 kV tube potential, the mean age, body mass index, heart rate, heart rate variability, and acquisition time variability(P>0.05). The minimum and maximum block acquisition time was shorter in group A than in group B (P<0.05). There were medium positive correlations between the acquisition time variability and heart rate variability in group A (r=0.457, P< 0.05) and B (r=0.506, P<0.05).The mean segmental score is 1.54 ± 0.75,1.52 ± 0.70, and 1.56 ± 0.73 in group A, B, and C, respectively. No significant differences were found in image quality between group A and B (P>0.05), and between group A and C (P>0.05). Image quality was better in group B than in group C (P<0.05). There was good agreement on image quality scores between the two reviewers (Kappa=0.762). There were 966 (98.2%),953 (98.7%), and 948 (98.1%) diagnostic segments in group A, B, and C, respectively. No significant differences were found in the ratio of diagnostic segments between group A and group B, between group A and group C, and between group B and group C (P> 0.05).The mean vessel score is 1.91 ± 0.87,1.88 ± 0.82, and 1.91 ± 0.85 in group A, B, and C, respectively. At all vessel level, no significant differences were found in image quality between group A and group B, and between group A and group C (P>0.05), but statistical difference was found between group B and group C (P<0.05). There were 95.5%(275/288),96.5%(278/288), and 95.5%(275/288) of diagnostic vessels in group A, B, and C, respectively. No significant differences were found in the ratio of diagnostic vessels among the three groups (P>0.05).At per-patient level, the mean subjective score is 2.58±0.75,2.60 ± 0.73, and 2.68 ± 0.77 in group A, B, and C, respectively. No significant differences were found in image quality between group A and group B, and between group A and group C (P> 0.05), but image quality is better in group B than in group C (P< 0.05). There were 88.9%(64/72),91.7%(66/72), and 87.5%(63/72) of diagnostic patients in group A, B, and C, respectively. No significant differences were found in the ratio of diagnostic patients among the three groups (P>0.05).The CT value in the root of AA in group B was slightly higher than in group C (P<0.05). No significantly differences were found in the other objective image quality parameters among the three groups (P>0.05). The radiation dose parameters in group A were lower than in group B (P<0.05).Study 2:There were no significant differences in the ratio of male, the number of scan blocks, and the acquisition windows, the mean age, body mass index, heart rate, and heart rate variability (P>0.05). There were 620 and 627 segments reviewed in group A and B, respectively. There was fair agreement on image quality scores between the two reviewers (Kappa=0.598). The mean segmental score in group B (1.68±0.73) was better than in group A (1.80±0.77) (P<0.05). There were 97.3% (603/620) and 97.6%(612/627) of diagnostic segments in group A and B, respectively. No significant differences were found in the ratio of diagnostic segments between the two groups (P>0.05).The mean vessel score in group A is 2.18 ± 0.85 and the score in group B is 2.06 ± 0.86. No significant difference was found between the two scores (P>0.05). At per-vessel level, only LAD score in group B (2.34 ± 0.77) was better than in group A (2.62 ± 0.70) (P<0.05). No significant difference was found in the subjective scores in RCA, LM and LCX between the two groups (P>0.05). There were 93.5%(187/200) and 92.5%(185/200) of diagnostic vessels in group A and B, respectively. No significant differences were found in the ratio of diagnostic vessels between the two groups (P>0.05).At per-patient level, the mean patient score in group A is 2.94 ± 0.65 and that in group B is 2.74 ± 0.80. No significant difference was found between the two scores (P>0.05). There were 82.0%(41/50) and 90.0%(40/50) of diagnostic patients in group A and B, respectively. No significant differences were found in the ratio of diagnostic patients between the two groups (P>0.05).The CT value and noise in group A are higher than in group B (P<0.05). The SNR and CNR in group A are lower than in group B (P<0.05). The radiation dose parameters in group A were lower than in group B (P<0.05).ConclusionProspectively ECG-triggered adaptive sequential CCTA with single-phase acquisition protocol in middle diastole, using the first generation of 2 x 64 slices dual source CT, is feasible for patients with low and stable heart rates. It can reduce radiation dose significantly while maintaining image quality and diagnostic performance compared with multi-phase protocol.Prospectively ECG-triggered adaptive sequential CCTA with an ultra-low tube potential of 70 kV, using the second generation of 2 x 128 slices dual source CT, is feasible for patients with a body mass index of< 26 kg/m2. It can reduce radiation dose significantly while maintaining similar diagnostic image quality compared with 80 kV protocol.