- Email: [email protected]
Application of Adaptive Statistical Iterative Reconstruction-V With Combination of 80 kV for Reducing Radiation Dose and Improving Image Quality in Renal Computed Tomography Angiography for Slim Patients
ARTICLE IN PRESS
Original Investigation
Application of Adaptive Statistical Iterative Reconstruction-V With Combination of 80 kV for Reducing Radiation Dose and Improving Image Quality in Renal Computed Tomography Angiography for Slim Patients D2X XZhanli Ren, D3X XMD#, D4X XXirong Zhang, D5X XMD#, D6X XZhijun Hu, D7X XMD, D8X XDou Li, D9X XMD, D10X XZhentang Liu, D1X XMD, D12X XDonghong Wei, D13X XMD, D14X XYongjun Jia, D15X XMD, D16X XNan Yu, D17X XMD, D18X XYong Yu, D19X XMD, D20X XYuxin Lei, D21X XMD, D2X XXiaoxia Chen, D23X XMD, D24X XChangyi Guo, D25X XMD, D26X XZhanliang Ren, D27X XMD, D28X XTaiping He, D29X XMD Objectives: To explore the application of adaptive statistical iterative reconstruction-V (ASIR-V) with combination of 80 kV for reducing radiation dose and improving image quality in renal computed tomography angiography (CTA) for slim patients compared with traditional filtered back projection (FBP) reconstruction using 120 kV. Methods: Eighty patients for renal CTA were prospectively enrolled and randomly divided into group A and group B. Group A used 120 kV and 600 mgI/kg contrast agent and FBP reconstruction, while group B used 80 kV and 350 mgI/kg contrast agent and both FBP and ASIR-V reconstruction from 10%ASIR-V to 100%ASIR-V with 10%ASIR-V interval. The CT values and SD values of the right renal artery and left renal artery were measured to calculate the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR). The image quality was subjectively scored by two experienced radiologists blindly using a five-point criterion. The contrast agent, volumetric CT dose index (CTDIvol), and dose length product in both groups were recorded and the effective radiation dose was calculated. Results: There were no significant difference in patient characteristics between two groups (p > 0.05). The CTDIvol, dose length product and effective radiation dose in group B were 59.0%, 65.0%, and 65.1% lower than those in group A, respectively (all p < 0.05), and the contrast agent in group B was 42.2% lower than that in group A (p < 0.05). In group B, with the increase of ASIR-V percentage, CT values showed no significant difference, SD values decreased gradually, SNR values and CNR values increased gradually. The CT values showed no statistically significant difference (p > 0.05) between two groups with different reconstructions. The SD values with 40%ASIR-V to 100%ASIR-V reconstruction in group B was significantly lower(p < 0.5), while the SNR values with 50% ASIR-V to 100% ASIR-V reconstruction and CNR values with 70%ASIR-V to 100%ASIR-V were significantly higher than those of group A with FBP reconstruction (p < 0.5). Two radiologists had excellent consistency in subjective scores of image quality for renal CTA (kappa >0.75, p < 0.05). The subjective scores with 60% ASIR-V to 90% ASIR-V in group B were significantly higher than those of FBP in group A (p < 0.5), of which 70%ASIR-V reconstruction obtained the highest subjective score for renal CTA. Conclusion: ASIR-V with combination of 80 kV can significantly reduce effective radiation dose (about 65.1%) and contrast agent (about 42.2%) and improve image quality in renal CTA for slim patients compared with traditional FBP reconstruction using 120 kV, and the 70% ASIR-V was the best reconstruction algorithm in 80 kV renal CTA. Advances in Knowledge: Using 80 kV with combination of ASIR-V can significantly reduce radiation dose and contrast agent dose as well as improve image quality in renal CTA for thin patients when compared with FBP using 120 kV. © 2019 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.
Acad Radiol 2019; &:19 From the Affiliated Hospital of Shaanxi University of Chinese Medicine, Weiyang western road- 2#, Xianyang, Shaanxi, China 712000 (Z.R., X.Z., Y.J., N.Y., Y.Y., Y.L., X.C., Z.R., T.H.); Shaanxi University of Chinese Medicine, Xianyang, Shaanxi, China (Z.R.); The Second Affiliated Hospital of Shaanxi University of Chinese Medicine, Xianyang, Shaanxi, China (Z.R., C.G.); Department of Medical Imaging, Chang'an Hospital, Xi'an, Shaanxi, China (Z.H., D.L., Z.L., D.W.). Received November 3, 2018; revised December 23, 2018; accepted December 24, 2018. Address correspondence to: Z. R.; T. H. e-mail: [email protected] # Co-first author: These authors contributed equally to this work. © 2019 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.acra.2018.12.021
1
ARTICLE IN PRESS REN ET AL
Academic Radiology, Vol &, No &&, && 2019
INTRODUCTION
MATERIALS AND METHODS
W
Patient Population
ith the rapid development and continuous updating of new computed tomography (CT) imaging techniques, computed tomography angiography (CTA), as a noninvasive angiography method, has the advantages of high spatial resolution and high density resolution (1), which plays an important role in the evaluation of abdominal vascular anatomy and vascular diseases. Renal CTA is the main imaging method for evaluating renal artery anatomy and diagnosing renal artery diseases, the better image quality for renal CTA depends on the higher iodine concentration in renal artery lumen, the better vascular contrast between renal artery and surrounding tissue and low image noise. However, improving the image quality of CTA by increasing contrast agent dose would result in a series of clinical concerns, such as contrast-induced nephropathy (2,3), radiation-induced injury (4). At present, the radiation dose is reduced by using low tube voltage (5), low tube current or automatic tube current modulation technique (6,7), and iterative reconstruction (8),of which low tube voltage can significantly improve contrast enhancement between renal artery and surrounding tissues, but the low radiation dose obtained by using low tube voltage can result in high image noise and poor image quality (9,10). Filtered back projection (FBP) is used as a traditional and standard CT reconstruction algorithm, but it is very sensitive to signal intensity and has the characteristics of high noise, obvious streaking artifacts, and poor lesion detectability at the condition of low radiation dose, which cannot produce high image quality and may not meet the requirement of clinical diagnosis at low radiation dose. In recent years, iterative reconstruction (IR) algorithms have been developed rapidly and used in CT application to ensure high image quality at low radiation dose for clinical diagnosis (11,12). One of the widely used IR algorithms is the adaptive statistical iterative reconstruction (ASIR), which could reduce 25%-40% radiation dose on the premise of guaranteeing image quality compared with FBP algorithm (13,14). However, the higher percentage of ASIR algorithm may result in waxy artifacts and decrease spatial resolution (15). Adaptive statistical iterative reconstruction-V (ASIR-V) algorithm, a recently developed and advanced iterative reconstruction algorithm, adopts more advanced noise model than ASIR reconstruction and introduces object model and physical model, which has greater potential to reduce image noise and reduce radiation dose than ASIR algorithm (16,17). Therefore, the purpose of this study was to explore the clinical application of ASIR-V algorithm with combination of low tube voltage at 80 kV for reducing radiation dose and contrast agent dose and improving image quality in renal CTA for slim patients in comparison with the conventional FBP reconstruction algorithm with standard 120 kV tube voltage.
2
This study was approved by the institutional review board of our hospital and a written informed consent was obtained from each participant prior to the study. Patients for renal CTA were prospectively enrolled in our hospital from April 2018 to July 2018. The inclusion criteria were as follows: patients older than 18 years old, patients with body mass index (BMI) 25 kg/m2, patients with clinical disease conditions requiring renal CTA examination, and patients being voluntary to participate in the study, normal renal CTA without renal artery diseases. The exclusion criteria were listed as follows: patients with allergy to iodine contrast agent, patients with known renal dysfunction or heart failure or liver dysfunction, and those patients with failed renal CTA examination. Eighty patients were enrolled in the study finally and randomly divided into group A and group B. Group A consisted of 26 males and 14 females, with an average age of 58.80 § 10.56 years (38-75 years) and BMI of 22.24 § 2.66 kg/m2, while group B consisted of 21 males and 19 females, with an average age of 56.78 § 16.73 years (42-75 years) and BMI of 22.52 § 2.92 kg/m2.
Scanning Technique
All patients were scanned on a Revolution CT scanner (GE Healthcare, Waukesha WI). The two groups of patients were scanned in feet-first supine position, with their hands placed on the top of the head, and thyroid and pelvis protected with lead aprons. The nonionic contrast agent Ultravist 370 (370 mgI/ml, Bayer Healthcare Ltd, Guangzhou, China) was used. Patients in group A were scanned with 120 kV tube voltage and 600 mgI/kg contrast agent dose while patients in group B were scanned with 80 kV tube voltage and 350 mgI/kg contrast agent dose. Contrast agent was injected via the median elbow vein by using a dual-tube power injector (Ulrich, German). The total injection time was fixed at 30 seconds, and the injection rate was obtained according to the formula that the total amount of contrast agent divided by the total injection time (injection rate = total contrast agent/30 seconds). The other scanning parameters of the two groups were kept consistent, automatic tube current modulation to achieve noise index of 10 HU, helical pitch with 0.992:1, rotation time of 0.5 second, detector width of 80 mm, coverage speed of 158.75 mm/s, image layer thickness and interval of 5 mm. The scanning area ranged from the level of bilateral rib-diaphragm angle to superior margin of bilateral iliac bone. The fixed threshold triggering technique was used in the arterial phase, and the monitoring region of interest (ROI) was placed in the abdominal aorta lumen at the level of cardio-diaphragmatic angle with a threshold value of 200 HU.
ARTICLE IN PRESS Academic Radiology, Vol &, No &&, && 2019
Image Reconstruction
After the CT data acquisition, the original data of group A were reconstructed with FBP algorithm while the original data of group B were reconstructed with FBP algorithm and ASIR-V algorithm from 10%ASIR-V to 100%ASIR-V with an interval of 10%ASIR-V. All the reconstructed images had a 1.25 mm slice thickness with window width of 350 HU and window level of 40 HU. Images Evaluation
Objective Image Evaluation The images reconstructed with FBP algorithm in group A and images reconstructed with FBP and ASIR-V algorithm with different percentages in group B were transferred to AW4.6 workstation for data measurement and image evaluation. ROI were placed on the maximum lumen center of the left renal artery and right renal artery 1 cm from the opening of the renal artery and the upper axial image adjacent to the maximum lumen center of renal artery and the lower axial image adjacent to the maximum lumen center of renal artery (three measurements for each renal artery) and the right erector spinae at the same level, the CT values and standard deviation (SD) of each ROI were recorded, and the mean values of the three ROIs were taken as the CT value and SD value of the renal arteries and erector spinae. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of renal artery were calculated with the SD values of the right erector spinae as background image noise. The SNR of renal artery was defined as the ratio of the CT value of renal artery divided by the SD value of renal artery (SNR = CT renal artery/SD renal artery), while the CNR of renal artery was defined as the difference of CT value between renal artery and erector spinae divided by the SD value of erector spinae (CNR = [CT renal artery ¡ CT erector spinae]/SD erector spinae). Subjective Image Evaluation For the subjective evaluation of image quality, the volume rendering and maximum intensity projection images were generated on the AW4.6 workstation, in addition to the original axial images. The image quality was scored subjectively and blindly on the axial images, volume rendering images and maximum intensity projection images in terms of vessel contrast, image artifacts, image noise, and diagnostic confidence through a comprehensive single subjective score by two radiologists with more than 10 years of working experience using a five-point scoring criterion. The
ASIR-V AND 80KV IN RENAL CTA
scoring criteria are listed in Table 1. In the five-point scoring scale, three points and more were considered to have clinically diagnostic image quality, two points and below did not meet the requirement for clinical diagnosis. Radiation Dose and Contrast Agent Estimation
To estimate the radiation dose, the volumetric CT dose index (CTDIvol) and dose length product (DLP) of the two groups were recorded and the effective dose (ED) was calculated using the formula ED = DLP £ k, of which k represents the radiation dose conversion factor with a value of 0.015 mSv/(mGy.cm) for the abdomen (18). The patient weight in the two groups was recorded, and the contrast agent dose was calculated according to the weight using the formula: contrast agent (g) = contrast agent dose (mgI/kg) £ weight (kg) £ 0.001. Statistical Analysis
All the measurements were analyzed by using SPSS22.0 statistical analysis software. Continuous data were expressed in the form of mean § SD. The difference of the patient characteristics (except for sex which was tested using x2 test), radiation dose and contrast agent between group A and group B were compared using the independent samples t test. The CT values, SD values, SNR values and CNR values of all reconstructed images was analyzed by one-way ANOVA and LSD-t test was used to compare every two reconstruction pairs. The subjective scores for all reconstructed images in terms of vessel contrast, artifacts, noise, and diagnostic confidence by two radiologists were compared with Kruskal-Wallis H test, and the consistency of subjective score between the two radiologists was analyzed using the kappa test, with kappa value 0.75 indicating an excellent consistency, 0.4< kappa value <0.75 indicating good consistency, and kappa value 0.4 indicating poor consistency. For all statistical tests, p < 0.05 being statistically significant. RESULTS Patient Characteristics, Radiation Dose, and Contrast Agent
The patient characteristics, radiation dose, and contrast agent of two groups were listed in Table 2. The patient characteristics including sex, age, weight, height, and BMI showed no statistically significant difference between the two groups
TABLE 1. The Subjective Score Criteria of Image Quality Scores
Image Quality
Vessel Contrast
Artifact
Noise
Diagnostic Confidence
5 4 3 2 1
Best Good Moderate Poor Worse
Very clear Clear Less clear Unclear Cannot be displayed
Very little Mild Moderate Obvious Severe
Very little Mild Moderate Obvious Severe
Fully diagnostic Good diagnostic Diagnostic Affecting diagnosis Nondiagnostic
3
ARTICLE IN PRESS Academic Radiology, Vol &, No &&, && 2019
REN ET AL
TABLE 2. The Comparison of Patients Characteristics, Radiation Dose and Contrast Agent Between Two Groups Parameters Patients characteristics Sex (Male/Female) Age (Year) Weight (kg) Height (cm) BMI (kg/㎡) Radiation dose CTDIvol (mGy) DLP (mGycm) ED (mSv) Contrast agent (g)
Group A
Group B
x2 value/t Value
p Value
26/14 58.80 § 10.56 63.25 § 6.27 168.90 § 6.57 22.24 § 2.66
21/19 56.78 § 16.73 62.63 § 8.48 166.75 § 4.60 22.52 § 2.92
1.289a 0.647b 0.375b 1.695b -0.438b
0.256 0.519 0.709 0.094 0.663
13.00 § 4.08 402.46 § 126.28 6.04 § 1.89 37.95 § 3.76
5.33 § 0.29 140.88 § 12.99 2.11 § 0.19 21.92 § 2.97
11.855b 13.032b 13.032b 21.162b
<0.001# <0.001# <0.001# <0.001#
a x2 value. b t value. # p < 0.05 with independent samples t test.
(p > 0.05). There were significant differences in CTDIvol, DLP, ED and contrast agent between group A and group B (all p < 0.001), the CTDIvol, DLP, ED in group B were 59.0%, 65.0%, and 65.1% lower than those in group A, respectively (all p < 0.001),and the contrast agent dose in group B was 42.2% lower than that in group A (p < 0.001). The relationship between CTDI and BMI in group A and group B were showed in Figure 3 ab.
Objective Image Evaluation
The comparison of CT values, SD values, SNR values, and CNR values between two groups with different reconstruction were listed in Table 3. There was no significant difference in CT values of the left renal artery and right renal artery between the two groups for all reconstructed images (p > 0.05), while the difference in SD values, SNR values, and CNR values between two groups with different reconstruction algorithms were statistically significant (p < 0.05). Compared with FBP reconstruction in group A, the SD value in group B with FBP reconstruction was significantly higher (p < 0.05), SNR value was slightly lower (p > 0.05) while CNR value was significantly lower (p < 0.05). With the increase of ASIR-V percentage in group B, CT values of renal artery showed no significant difference, while the SD values gradually decreased, and SNR values and CNR values gradually increased. The SD values in group B with 40% ASIR-V to 100% ASIR-V reconstructions were statistically lower than that in group A (p < 0.05), SNR values with 50% ASIR-V to 100% ASIR-V reconstruction and CNR values with 70% ASIR-V to 100% ASIR-V reconstruction were significantly higher than those in group A (p < 0.05). Subjective Image Evaluation
The subjective scores of renal CTA are listed in Table 4, and subjective scores of all the reconstructed images with FBP algorithm and ASIR-V reconstruction were more than three points. The 4
subjective scores of reconstructed images in terms of vessel contrast, artifact, noise, and diagnostic confidence were well consistent between the two radiologists (kappa value >0.75, p < 0.05). There were significant differences in the subjective scores of in terms of vessel contrast, artifacts, noise, and diagnostic confidence between the two groups with different reconstruction algorithms (p < 0.05). Compared with FBP reconstruction in group A, the scores of artifacts and noise and diagnostic confidence with FBP reconstruction in group B were significantly lower (p < 0.05), while the score of vessel contrast with FBP reconstruction in group B was slightly lower (p > 0.05). The subjective scores of renal CTA in group B with 60% ASIR-V to 90% 0X3D X ASIR-V reconstruction in terms of vessel contrast, artifacts, noise, and diagnostic confidence were significantly higher than those of FBP reconstruction in group A (p < 0.05), of which the 70%ASIR-V reconstruction achieved the highest comprehensive subjective scores (Fig. 1 and 2).
DISCUSSION RenalCTA, as a fast and noninvasive imaging method, has the characteristic of excellent spatial resolution and high density resolution, which brings the clear vessel contrast between renal artery and its surrounding tissue and help to display the anatomy and abnormality of renal artery clearly. Nevertheless, the main disadvantages of CTA are ionizing radiation exposure and potential contrast-induced nephropathy (19). Therefore, it is very important to reduce radiation dose and contrast agent dose in the process of renal CTA while meeting clinical diagnostic image quality. FBP, as a traditional and standard method for CT image reconstruction, is difficult to find a balance between high image quality and low radiation dose. IR is a new CT reconstruction algorithm, which has greater potential in reducing image noise, suppressing artifacts, and reducing radiation dose compared with traditional FBP algorithm. ASIR, being one of the early IR algorithms that based on noise model, has been widely used in clinical work, which not only helps to reduce radiation dose
Reconstruction Algorithm
CT Value (HU)
Noise (SD Value)
Signal to Noise Ratio (SNR)
Contrast to Noise Ratio (CNR)
Right Renal Artery Left Renal Artery Right Renal Artery Left Renal Artery Right Renal Artery Left Renal Artery Right Renal Artery Left Renal Artery 271.78 § 66.55 282.79 § 54.25 282.79 § 54.25 282.04 § 54.29 283.03 § 53.08 283.46 § 53.05 283.30 § 53.14 282.72 § 53.22 282.43 § 53.37 281.88 § 53.51 281.40 § 53.68 280.16 § 51.42 0.138 0.986
273.32 § 65.28 289.94 § 49.81 289.53 § 49.82 289.11 § 49.95 288.29 § 50.67 290.01 § 49.57 289.90 § 49.97 288.75 § 51.33 289.43 § 50.51 289.36 § 50.60 289.06 § 50.80 288.76 § 51.31 0.320 0.982
23.07 § 3.07 29.78 § 6.42* 27.41 § 6.09* 24.81 § 5.38 22.82 § 5.10 20.49 § 4.15* 18.38 § 3.97* 16.15 § 3.69* 14.04 § 3.58* 12.01 § 3.47* 10.23 § 3.48* 8.62 § 3.14* 96.206 <0.001|
23.13 § 3.63 29.93 § 4.48* 27.71 § 4.22* 25.08 § 3.89* 23.12 § 3.87 20.74 § 3.06* 18.63 § 3.08* 16.27 § 3.19* 14.21 § 3.40* 12.04 § 3.73* 10.34 § 4.04* 8.41 § 4.57* 135.155 |
11.99 § 3.38 9.93 § 2.75 10.79 § 2.88 11.85 § 3.20 12.94 § 3.21 14.28 § 3.47 15.95 § 3.80* 18.15 § 4.27* 21.02 § 5.13* 24.96 § 6.94* 30.47 § 11.23* 37.11 § 18.10* 56.922 <0.001|
12.07 § 3.30 9.81 § 1.72 10.59 § 1.85 11.69 § 2.08 12.67 § 2.28 14.16 § 2.45 15.80 § 2.75* 18.14 § 3.41* 21.04 § 4.25* 25.52 § 6.62* 31.00 § 10.42* 42.27 § 20.18* 73.636 <0.001|
12.26 § 4.44 7.40 § 2.03* 7.97 § 2.22* 8.81 § 2.53* 9.60 § 2.73* 10.75 § 3.17 11.96 § 3.69 13.67 § 4.43 15.79 § 5.51* 18.48 § 7.46* 21.83 § 10.36* 25.48 § 11.33* 38.464 <0.001|
12.34 § 4.32 7.62 § 1.90* 8.21 § 2.07* 9.09 § 2.35* 9.82 § 2.61* 11.05 § 2.97 12.28 § 3.46 13.99 § 4.19 16.22 § 5.17* 19.01 § 7.06* 22.45 § 9.87* 26.25 § 11.08* 44.563 <0.001|
* p < 0.05 with FBP algorithm and ASIR-V algorithm in group B compared with FBP algorithm in group A. | p < 0.05 with one-way ANOVA.
TABLE 4. The Comparison of Subjective Scores of Renal CTA With Different Reconstruction Algorithms Reconstruction Algorithm
Artifact
Noise
Diagnostic Confidence
Radiologist 1
Radiologist 2
Radiologist 1
Radiologist 2
Radiologist 1
Radiologist 2
Radiologist 1
Radiologist 2
4.05 § 0.78 3.95 § 0.75 3.98 § 0.77 4.00 § 0.82 4.03 § 0.70 4.18 § 0.71 4.40 § 0.59* 4.78 § 0.42* 4.93 § 0.27* 4.73 § 0.45* 4.45 § 0.55* 4.28 § 0.45 69.064 <0.001❀
4.10 § 0.74 3.90 § 0.78 3.93 § 0.80 4.03 § 0.79 4.08 § 0.80 4.23 § 0.62 4.43 § 0.64* 4.70 § 0.46* 4.90 § 0.30* 4.88 § 0.34* 4.50 § 0.51* 4.33 § 0.47 69.199 <0.001❀
3.90 § 0.69 3.40 § 0.49* 3.52 § 0.50* 3.60 § 0.52* 3.93 § 0.57 4.10 § 0.55 4.28 § 0.51* 4.50 § 0.56* 4.80 § 0.41* 4.69 § 0.46* 4.63 § 0.49* 4.33 § 0.47* 70.640 <0.001❀
3.93 § 0.73 3.45 § 0.50* 3.58 § 0.51* 3.73 § 0.45 4.03 § 0.66 4.13 § 0.52 4.38 § 0.54* 4.63 § 0.54* 4.78 § 0.42* 4.73 § 0.45* 4.55 § 0.50* 4.35 § 0.48* 64.740 <0.001❀
3.95 § 0.69 3.38 § 0.49* 3.55 § 0.50* 3.88 § 0.61* 3.88 § 0.61 4.12 § 0.56 4.25 § 0.50* 4.43 § 0.55* 4.70 § 0.46* 4.75 § 0.44* 4.85 § 0.36* 4.93 § 0.27* 106.941 <0.001❀
3.92 § 0.70 3.45 § 0.51* 3.53 § 0.51* 3.68 § 0.47* 3.98 § 0.70 4.15 § 0.51 4.35 § 0.53* 4.55 § 0.55* 4.68 § 0.47* 4.83 § 0.39* 4.85 § 0.36* 4.90 § 0.30* 98.650 <0.001❀
4.25 § 0.63 3.68 § 0.47* 3.70 § 0.46* 3.83 § 0.39* 3.88 § 0.46* 4.10 § 0.63 4.40 § 0.49 4.58 § 0.50* 4.89 § 0.30* 4.73 § 0.45* 4.65 § 0.48* 4.34 § 0.48 60.644 <0.001❀
4.30 § 0.56 3.70 § 0.46* 3.78 § 0.42* 3.88 § 0.34* 3.98 § 0.66* 4.18 § 0.55 4.45 § 0.50 4.65 § 0.48* 4.82 § 0.42* 4.73 § 0.45* 4.58 § 0.50* 4.50 § 0.51 40.265 <0.001❀
* p < 0.05 with FBP algorithm and ASIR-V algorithm in group B compared with FBP algorithm in group A. ❀ p < 0.05 with Kruskal-Wallis H test.
ASIR-V AND 80KV IN RENAL CTA
120kV-FBP 80kV-FBP 10%ASIR-V 20%ASIR-V 30%ASIR-V 40%ASIR-V 50%ASIR-V 60%ASIR-V 70%ASIR-V 80%ASIR-V 90%ASIR-V 100%ASIR-V Z value P value
Vessel Contrast
ARTICLE IN PRESS
120kV-FBP 80kV-FBP 10%ASIR-V 20%ASIR-V 30%ASIR-V 40%ASIR-V 50%ASIR-V 60%ASIR-V 70%ASIR-V 80%ASIR-V 90%ASIR-V 100%ASIR-V F value P value
Academic Radiology, Vol &, No &&, && 2019
TABLE 3. The Comparison of Objective Parameters of Renal CTA With Different Reconstruction Algorithms
ARTICLE IN PRESS REN ET AL
Academic Radiology, Vol &, No &&, && 2019
Figure 1. (a) A women aged 59 years with small renal cyst in right kidney, the patient was scanned with 120 kV and 600 mgI/kg and the images was reconstructed with FBP algorithm, volume render (a) showed mild noise and the subjective scores were fourpoint by two radiologists. (be), A 61year-old man with small renal cyst in left kidney, the patient was performed with 80 kV and 350 mgI/kg and the images were reconstructed with FBP (b), 40%ASIR-V (c), 70%ASIR-V(d), 100%ASIR-V (e), the subjective scores for volume render (be) were 3-point, 4-point, 5-point, 4-point by two radiologists, respectively.
and but also helps to provide better image quality. However, ASIR algorithm with higher percentage has tendency to over smooth the image quality, which results in artificial textures or blotchy appearance (20). Model-based iterative reconstruction (MBIR) is another advanced iterative reconstruction algorithm, which relies on more complex and accurate models than ASIR algorithm (21), and which has greater potential to reduce image noise and improve spatial resolution than ASIR algorithm. However, MBIR algorithm requires much longer time for image reconstruction (22). In recent years, an improved ASIR-V, as a new generation IR algorithm, has been developed to balance the model complexness and reconstruction time. ASIR-V algorithm discards optical model and adopts more advanced noise model with the addition of object modeling and physics modeling to balance image noise and spatial resolution when compared with ASIR, which provides better image quality and lower radiation dose than ASIR reconstruction and shorter reconstruction time than MBIR (23). In this study, reducing the tube voltage from 120 kV to 80 kV realized the reduction of radiation dose by 65.1% and contrast agent dose reduction by 42.2%, which was related to that the radiation dose was directly proportional to the square of tube voltage (24), that is to say, slight change in tube voltage would cause large change in radiation dose. The lower tube voltage was closer to the k edge of iodine (33.2 keV) (25), increasing of photoelectric effect and the reducing Compton scattering, and increasing the X-ray attenuation of iodine, which improves the CT values of renal CTA and the 6
vessel contrast between renal artery and surrounding tissues. The increased attenuation of iodine was used to reduce the contrast agent dose in renal CTA in our study. Our results were similar with that of Nagayama Y (26) who reported that the radiation dose decreased by 33% and 65%, respectively when reducing tube voltage from 120 kV to 100 kV or 80 kV, and that of Thor D (27) who reported contrast agent decrease by 40%-44% by using 80 kV. All these studies indicated that the radiation dose and contrast agent could be significantly reduced by using lower tube voltage. The reduction of radiation dose in our study was much greater than that of Nakaura T (28) who reported that the radiation dose was only reduced by 20% via reducing tube voltage from 120 kV to 80 kV, and that of Del Gaizo A J (29) who discovered that using 80 kV could achieve a 30% reduction of radiation dose. The reduction of contrast dose in our study was also much greater than that of Feng C (30) who confirmed that contrast agent was reduced by 26.1% using 80 kV or 100 kV according to different BMI, all the researches performed by Nakaura T, Del Gaizo A J, and Feng C about reducing radiation dose and contrast agent was lower than the results of our study, which may be related to the differences of patient population in different studies, different types or injection rates of contrast agent, and differences caused by different scanners or scanning protocols. For comparison of FBP reconstruction between 120 kV and 80 kV, the results showed that SD values of 80 kV with FBP reconstruction was slightly higher than that of 120 kV, while SNR values and CNR values of 80 kV with FBP
ARTICLE IN PRESS Academic Radiology, Vol &, No &&, && 2019
Figure 2. (a) 57yearold man with 120 kV and FBP algorithm, the subjective scores of maximum intensity projection were fourpoint by two radiologists. (be), 56yearold woman with 80 kV, the images were reconstructed with FBP (b), 40%ASIR-V (c), 70%ASIR-V (D), 100%ASIR-V (e), the subjective scores of maximum intensity projection (be) were 3-point, 4-point, 5-point, 4-point by two radiologists, respectively.
reconstruction was lower than those of 120 kV. In addition, the subjective scores of 80 kV with FBP reconstruction in vessel contrast, artifact, noise, and diagnostic confidence were
ASIR-V AND 80KV IN RENAL CTA
significantly lower than those of 120 kV, which indicated that the reduction of radiation dose by using lower tube voltage would lead to the increase of image noise and the decrease of image quality while maintaining the consistency of other scanning parameters. With the increase of ASIR-V reconstruction percentage in group B, the SD values decreased gradually, and the SNR values and CNR values increased gradually within a certain range from 10%ASIR-V to 70%ASIR-V, indicating that ASIR-V reconstruction could significantly reduce image noise and improve image quality. Compared to the FBP reconstructions in group A with 120 kVp, the SD values with 70% ASIR-V to 100% ASIR-V reconstruction in group B was lower, SNR values and CNR values were significantly higher, which indicated that choosing higher percentage of ASIR-V reconstruction in 80 kV renal CTA could achieve better image quality than FBP reconstructions in 120 kVp. With the increase of ASIR-V percentage in group B, the subjective score of image noise increased gradually from 10%ASIR-V to 100%ASIR-V reconstruction, while the subjective score in vessel contrast, artifacts ,and diagnostic confidence gradually increased from 10%ASIR-V to 70%ASIR-V reconstruction, and then decreased gradually when the ASIR-V reconstruction percentage exceeding 70%. Namely, the 70%ASIR-V reconstruction algorithm could achieve the highest subjective score of overall image quality in 80 kV renal CTA when compared to FBP reconstruction in 120 kV and other ASIR-V reconstruction percentage in 80 kV. Therefore, 70%ASIR-V reconstruction algorithm could achieve the best renal CTA image quality combining subjective evaluation with objective evaluation. Our results showed that lower percentage ASIR-V reconstruction would limit the ability of suppressing image noise and higher percentage ASIR-V reconstruction would result in waxy artifacts or blocky appearance to affect the vessel contrast. However, the research carried by Benz D C (31) reported that 100%ASIR-V reconstruction produced
Figure 3. (ab) The Relationship of CTDI and patient BMI in group A (a) and group B (b).
7
ARTICLE IN PRESS Academic Radiology, Vol &, No &&, && 2019
REN ET AL
substantial noise reduction and improved image quality in lowdose coronary computed tomography angiography, while the research performed by Pontone G (32) discovered that 60%ASIR-V reconstruction provided the optimal balance between image objective parameters and image quality. The results of our study demonstrated the inconsistency with the researches performed by Benz D C and Pontone G, which could be explained by that our study focused on renal CTA and the research by Benz and Pontone focused on coronary artery, and the contrast agent concentration and injection rate would also affect the results of both objective and subjective image quality evaluation (33). This study has some limitations. Firstly, the sample size of our study was small; secondly, the patients population were randomly divided into two groups, which was not paired according to individual differences; thirdly, only normal renal CTA were included into the study population and renal artery disease was not assessed; finally, all the patients evaluated in our study had with small BMI values and patients with larger BMI values were not concluded in the study due to the population characteristics in China, this might introduce bias in our study. Therefore, in the future research work, we need to further expand the sample size to include patients with large BMI values, and the patient population will be paired to divide into groups according to the height, weight and age to reduce the impact of individual differences on the results of the study, and the researches about renal CTA using 80 kV will recruit patients with renal artery diseases. In conclusion, using 80 kV tube voltage with combination of ASIR-V algorithm can significantly reduce radiation dose and contrast agent dose as well as improve image quality in renal CTA for thin patients when compared to FBP using 120 kV. Compared to FBP using 120 kV or 80 kV and other ASIR-V reconstruction percentage, the 70% ASIR-V reconstruction percentage was the best reconstruction algorithm in 80 kV renal CTA for slim patients to achieve the optimal balance between radiation dose and image quality, which helps to provide support for further reduction of radiation dose and contrast agent dose in renal CTA as well as cost saving, and has potential to reduce clinical radiation exposure injury and contrastinduced nephropathy. ACKNOWLEDGMENTS The authors wish to thank Dr. Gu Jun for guidance and suggestions in this paper and Dr. Li Jianying for the modifying and editing the article. REFERENCES 1. 2.
3. 4.
8
Xu Y, Yu S, Zhang L, et al. Application value of iterative reconstruction with CTA to intractable headache patients. Exp Ther Med 2018; 16:603–608. Marenzi G, Assanelli E, Campodonico J, et al. Contrast volume during primary percutaneous coronary intervention and subsequent contrastinduced nephropathy and mortality. Ann Intern Med 2009; 150:170–177. Stacul, Fulvio, Claudio, et al., et al. Subclinical acute kidney injury (AKI) due to iodine-based contrast media. Eur Radiol. 2013; 23:319–323. Piechowiak EI, Peter JF, Kleb B, et al. Intravenous iodinated contrast agents amplify DNA radiation damage at CT. Radiology 2015; 275:692–697.
5.
6.
7. 8.
9.
10.
11. 12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
Raman SP, Mahesh M, Blasko RV, et al. CT scan parameters and radiation dose: practical advice for radiologists. J Am Coll Radiol 2013; 10:840–846. Nijhof WH, Baltussen EJ, Kant IM, et al. Low-dose CT angiography of the abdominal aorta and reduced contrast medium volume: assessment of image quality and radiation dose. Clin Radiol 2016; 71:64–73. Mayosmith WW, Hara AK, Mahesh M, et al. How I do it: managing radiation dose in CT. Radiology 2014; 273:657–672. Yan C, Xu J, Liang C, et al. Radiation Dose Reduction by Using CT with iterative model reconstruction in patients with pulmonary invasive fungal infection. Radiology 2018; 288:285–292. Iyama Y, Nakaura T, Yokoyama K, et al. Low-contrast and low-radiation dose protocol in cardiac computed tomography: usefulness of low tube voltage and knowledge-based iterative model reconstruction algorithm. J Comput Assist Tomogr 2016; 40:941–947. Chang KJ, Collins S, Li B, et al. Optimizing CT technique to reduce radiation dose: effect of changes in kVp, iterative reconstruction, and noise index on dose and noise in a human cadaver. Radiol Phys Technol 2017; 10:180–188. Geyer LL, Schoepf UJ, Meinel FG, et al. State of the art: iterative CT reconstruction techniques. Radiology 2015; 276:339–357. Patino M, Fuentes JM, Singh S, et al. Iterative reconstruction techniques in abdominopelvic CT: technical concepts and clinical implementation. Am J Roentgenol 2015; 205:W19–W31. Mitsumori LM, Shuman WP, Busey JM, et al. Adaptive statistical iterative reconstruction versus filtered back projection in the same patient: 64 channel liver CT image quality and patient radiation dose. Eur Radiol 2012; 22:138–143. Pickhardt PJ, Lubner MG, Kim DH, et al. Abdominal CT with modelbased iterative reconstruction (MBIR):initial results of a prospective trial comparing ultralow-dose with standard-dose imaging. Am J Roentgenol 2012; 199:1266–1274. Qi LP, Li Y, Tang L, et al. Evaluation of dose reduction and image quality in chest CT using adaptive statistical iterative reconstruction with the same group of patients. Br J Radiol 2012; 85:906–911. Marco PD, Origgi D. New adaptive statistical iterative reconstruction ASiRV: assessment of noise performance in comparison to ASiR[J]. J Appl Clin Med Phys 2018; 19:275–286. Kwon H, Cho J, Oh J, et al. The adaptive statistical iterative reconstruction-V technique for radiation dose reduction in abdominal CT: comparison with the adaptive statistical iterative reconstruction technique. Br J Radiol 2015; 88:20150463. Saman R, Co SJ, Simon B. Reduced kilovoltage in computed tomography-guided intervention in a community hospital: effect on the radiation dose. Can Assoc Radiol J 2014; 65:345–351. Pang L, Zhao Y, Dong H, et al. High-Pitch Dual-Source Computed Tomography Renal Angiography Comparison With Conventional Low-Pitch Computed Tomography Angiography: Image Quality, Contrast Medium Volume, and Radiation Dose. J Comput Assist Tomogr 2015; 39:737–740. Kim HG, Yong EC, Lee YH, et al. Quantitative analysis of the effect of iterative reconstruction using a phantom: determining the appropriate blending percentage. Yonsei Med J 2015; 56:253–261. Kim JH, Choo KS, Moon TY, et al. Comparison of the image qualities of filtered back-projection, adaptive statistical iterative reconstruction, and model-based iterative reconstruction for CT venography at 80 kVp. Eur Radiol 2016; 26:2055–2063. Gatti M, Marchisio F, Fronda M, et al. Adaptive statistical iterative reconstruction-V versus adaptive statistical iterative reconstruction: impact on dose reduction and image quality in body computed tomography. J Comput Assist Tomogr 2018; 42:191–196. Lim K, Kwon H, Cho J, et al. Initial phantom study comparing image quality in computed tomography using adaptive statistical iterative reconstruction and new adaptive statistical iterative reconstruction v. J Comput Assist Tomogr 2015; 39:443. Ippolito D, Talei FC, Fior D, et al. Low kV settings CT angiography (CTA) with low dose contrast medium volume protocol in the assessment of thoracic and abdominal aorta disease: a feasibility study. Br J Radiol 2015; 88:20140140. Hou P, Feng X, Liu J, et al. Low tube voltage and iterative model reconstruction in follow-up CT angiography after thoracic endovascular aortic repair: ultra-low radiation exposure and contrast medium dose. Acad Radiol 2018; 25:494–501. Nagayama Y, Oda S, Nakaura T. Radiation dose reduction at pediatric CT: use of low tube voltage and iterative reconstruction. Radio graphics 2018; 38:1421–1440.
ARTICLE IN PRESS Academic Radiology, Vol &, No &&, && 2019
Thor D, Brismar TB, Fischer MA. Low tube voltage dual source computed tomography to reduce contrast media doses in adult abdomen examinations: a phantom study. Med Phys 2015; 42:5100– 5109. 28. Takeshi N, Kazuo A, Natsuki M, et al. Abdominal dynamic CT in patients with renal dysfunction: contrast agent dose reduction with low tube voltage and high tube current-time product settings at 256-detector row CT. Radiology 2011; 261:467–476. 29. Del Gaizo AJ, Fletcher JG, Yu L, et al. Reducing radiation dose in CT enterography. Radiographics 2013; 33:1109–1124. 30. Feng C, Zhu D, Zou X, et al. The combination of a reduction in contrast agent dose with low tube voltage and an adaptive statistical iterative
ASIR-V AND 80KV IN RENAL CTA
27.
31.
32.
33.
reconstruction algorithm in CT enterography: effects on image quality and radiation dose. Medicine 2018; 97:e0151. €ni C, Mikulicic F, et al. Adaptive statistical iterative reconBenz DC, Gra struction-V: impact on image quality in ultralow-dose coronary computed tomography angiography. J Comput Assist Tomogr 2016; 40:958–963. Pontone G, Muscogiuri G, Andreini D, et al. Impact of a new adaptive statistical iterative reconstruction (ASIR)-V algorithm on image quality in coronary computed tomography angiography. Acad Radiol 2018. pii: S1076-6332 (18)30094-1. Schwarz F, Grandl K, Arnoldi A, et al. Lowering radiation exposure in CT angiography using automated tube potential selection and optimized iodine delivery rate. AJR Am J Roentgenol 2013; 200:W628–W634.
9
Original Investigation
Application of Adaptive Statistical Iterative Reconstruction-V With Combination of 80 kV for Reducing Radiation Dose and Improving Image Quality in Renal Computed Tomography Angiography for Slim Patients D2X XZhanli Ren, D3X XMD#, D4X XXirong Zhang, D5X XMD#, D6X XZhijun Hu, D7X XMD, D8X XDou Li, D9X XMD, D10X XZhentang Liu, D1X XMD, D12X XDonghong Wei, D13X XMD, D14X XYongjun Jia, D15X XMD, D16X XNan Yu, D17X XMD, D18X XYong Yu, D19X XMD, D20X XYuxin Lei, D21X XMD, D2X XXiaoxia Chen, D23X XMD, D24X XChangyi Guo, D25X XMD, D26X XZhanliang Ren, D27X XMD, D28X XTaiping He, D29X XMD Objectives: To explore the application of adaptive statistical iterative reconstruction-V (ASIR-V) with combination of 80 kV for reducing radiation dose and improving image quality in renal computed tomography angiography (CTA) for slim patients compared with traditional filtered back projection (FBP) reconstruction using 120 kV. Methods: Eighty patients for renal CTA were prospectively enrolled and randomly divided into group A and group B. Group A used 120 kV and 600 mgI/kg contrast agent and FBP reconstruction, while group B used 80 kV and 350 mgI/kg contrast agent and both FBP and ASIR-V reconstruction from 10%ASIR-V to 100%ASIR-V with 10%ASIR-V interval. The CT values and SD values of the right renal artery and left renal artery were measured to calculate the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR). The image quality was subjectively scored by two experienced radiologists blindly using a five-point criterion. The contrast agent, volumetric CT dose index (CTDIvol), and dose length product in both groups were recorded and the effective radiation dose was calculated. Results: There were no significant difference in patient characteristics between two groups (p > 0.05). The CTDIvol, dose length product and effective radiation dose in group B were 59.0%, 65.0%, and 65.1% lower than those in group A, respectively (all p < 0.05), and the contrast agent in group B was 42.2% lower than that in group A (p < 0.05). In group B, with the increase of ASIR-V percentage, CT values showed no significant difference, SD values decreased gradually, SNR values and CNR values increased gradually. The CT values showed no statistically significant difference (p > 0.05) between two groups with different reconstructions. The SD values with 40%ASIR-V to 100%ASIR-V reconstruction in group B was significantly lower(p < 0.5), while the SNR values with 50% ASIR-V to 100% ASIR-V reconstruction and CNR values with 70%ASIR-V to 100%ASIR-V were significantly higher than those of group A with FBP reconstruction (p < 0.5). Two radiologists had excellent consistency in subjective scores of image quality for renal CTA (kappa >0.75, p < 0.05). The subjective scores with 60% ASIR-V to 90% ASIR-V in group B were significantly higher than those of FBP in group A (p < 0.5), of which 70%ASIR-V reconstruction obtained the highest subjective score for renal CTA. Conclusion: ASIR-V with combination of 80 kV can significantly reduce effective radiation dose (about 65.1%) and contrast agent (about 42.2%) and improve image quality in renal CTA for slim patients compared with traditional FBP reconstruction using 120 kV, and the 70% ASIR-V was the best reconstruction algorithm in 80 kV renal CTA. Advances in Knowledge: Using 80 kV with combination of ASIR-V can significantly reduce radiation dose and contrast agent dose as well as improve image quality in renal CTA for thin patients when compared with FBP using 120 kV. © 2019 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.
Acad Radiol 2019; &:19 From the Affiliated Hospital of Shaanxi University of Chinese Medicine, Weiyang western road- 2#, Xianyang, Shaanxi, China 712000 (Z.R., X.Z., Y.J., N.Y., Y.Y., Y.L., X.C., Z.R., T.H.); Shaanxi University of Chinese Medicine, Xianyang, Shaanxi, China (Z.R.); The Second Affiliated Hospital of Shaanxi University of Chinese Medicine, Xianyang, Shaanxi, China (Z.R., C.G.); Department of Medical Imaging, Chang'an Hospital, Xi'an, Shaanxi, China (Z.H., D.L., Z.L., D.W.). Received November 3, 2018; revised December 23, 2018; accepted December 24, 2018. Address correspondence to: Z. R.; T. H. e-mail: [email protected] # Co-first author: These authors contributed equally to this work. © 2019 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.acra.2018.12.021
1
ARTICLE IN PRESS REN ET AL
Academic Radiology, Vol &, No &&, && 2019
INTRODUCTION
MATERIALS AND METHODS
W
Patient Population
ith the rapid development and continuous updating of new computed tomography (CT) imaging techniques, computed tomography angiography (CTA), as a noninvasive angiography method, has the advantages of high spatial resolution and high density resolution (1), which plays an important role in the evaluation of abdominal vascular anatomy and vascular diseases. Renal CTA is the main imaging method for evaluating renal artery anatomy and diagnosing renal artery diseases, the better image quality for renal CTA depends on the higher iodine concentration in renal artery lumen, the better vascular contrast between renal artery and surrounding tissue and low image noise. However, improving the image quality of CTA by increasing contrast agent dose would result in a series of clinical concerns, such as contrast-induced nephropathy (2,3), radiation-induced injury (4). At present, the radiation dose is reduced by using low tube voltage (5), low tube current or automatic tube current modulation technique (6,7), and iterative reconstruction (8),of which low tube voltage can significantly improve contrast enhancement between renal artery and surrounding tissues, but the low radiation dose obtained by using low tube voltage can result in high image noise and poor image quality (9,10). Filtered back projection (FBP) is used as a traditional and standard CT reconstruction algorithm, but it is very sensitive to signal intensity and has the characteristics of high noise, obvious streaking artifacts, and poor lesion detectability at the condition of low radiation dose, which cannot produce high image quality and may not meet the requirement of clinical diagnosis at low radiation dose. In recent years, iterative reconstruction (IR) algorithms have been developed rapidly and used in CT application to ensure high image quality at low radiation dose for clinical diagnosis (11,12). One of the widely used IR algorithms is the adaptive statistical iterative reconstruction (ASIR), which could reduce 25%-40% radiation dose on the premise of guaranteeing image quality compared with FBP algorithm (13,14). However, the higher percentage of ASIR algorithm may result in waxy artifacts and decrease spatial resolution (15). Adaptive statistical iterative reconstruction-V (ASIR-V) algorithm, a recently developed and advanced iterative reconstruction algorithm, adopts more advanced noise model than ASIR reconstruction and introduces object model and physical model, which has greater potential to reduce image noise and reduce radiation dose than ASIR algorithm (16,17). Therefore, the purpose of this study was to explore the clinical application of ASIR-V algorithm with combination of low tube voltage at 80 kV for reducing radiation dose and contrast agent dose and improving image quality in renal CTA for slim patients in comparison with the conventional FBP reconstruction algorithm with standard 120 kV tube voltage.
2
This study was approved by the institutional review board of our hospital and a written informed consent was obtained from each participant prior to the study. Patients for renal CTA were prospectively enrolled in our hospital from April 2018 to July 2018. The inclusion criteria were as follows: patients older than 18 years old, patients with body mass index (BMI) 25 kg/m2, patients with clinical disease conditions requiring renal CTA examination, and patients being voluntary to participate in the study, normal renal CTA without renal artery diseases. The exclusion criteria were listed as follows: patients with allergy to iodine contrast agent, patients with known renal dysfunction or heart failure or liver dysfunction, and those patients with failed renal CTA examination. Eighty patients were enrolled in the study finally and randomly divided into group A and group B. Group A consisted of 26 males and 14 females, with an average age of 58.80 § 10.56 years (38-75 years) and BMI of 22.24 § 2.66 kg/m2, while group B consisted of 21 males and 19 females, with an average age of 56.78 § 16.73 years (42-75 years) and BMI of 22.52 § 2.92 kg/m2.
Scanning Technique
All patients were scanned on a Revolution CT scanner (GE Healthcare, Waukesha WI). The two groups of patients were scanned in feet-first supine position, with their hands placed on the top of the head, and thyroid and pelvis protected with lead aprons. The nonionic contrast agent Ultravist 370 (370 mgI/ml, Bayer Healthcare Ltd, Guangzhou, China) was used. Patients in group A were scanned with 120 kV tube voltage and 600 mgI/kg contrast agent dose while patients in group B were scanned with 80 kV tube voltage and 350 mgI/kg contrast agent dose. Contrast agent was injected via the median elbow vein by using a dual-tube power injector (Ulrich, German). The total injection time was fixed at 30 seconds, and the injection rate was obtained according to the formula that the total amount of contrast agent divided by the total injection time (injection rate = total contrast agent/30 seconds). The other scanning parameters of the two groups were kept consistent, automatic tube current modulation to achieve noise index of 10 HU, helical pitch with 0.992:1, rotation time of 0.5 second, detector width of 80 mm, coverage speed of 158.75 mm/s, image layer thickness and interval of 5 mm. The scanning area ranged from the level of bilateral rib-diaphragm angle to superior margin of bilateral iliac bone. The fixed threshold triggering technique was used in the arterial phase, and the monitoring region of interest (ROI) was placed in the abdominal aorta lumen at the level of cardio-diaphragmatic angle with a threshold value of 200 HU.
ARTICLE IN PRESS Academic Radiology, Vol &, No &&, && 2019
Image Reconstruction
After the CT data acquisition, the original data of group A were reconstructed with FBP algorithm while the original data of group B were reconstructed with FBP algorithm and ASIR-V algorithm from 10%ASIR-V to 100%ASIR-V with an interval of 10%ASIR-V. All the reconstructed images had a 1.25 mm slice thickness with window width of 350 HU and window level of 40 HU. Images Evaluation
Objective Image Evaluation The images reconstructed with FBP algorithm in group A and images reconstructed with FBP and ASIR-V algorithm with different percentages in group B were transferred to AW4.6 workstation for data measurement and image evaluation. ROI were placed on the maximum lumen center of the left renal artery and right renal artery 1 cm from the opening of the renal artery and the upper axial image adjacent to the maximum lumen center of renal artery and the lower axial image adjacent to the maximum lumen center of renal artery (three measurements for each renal artery) and the right erector spinae at the same level, the CT values and standard deviation (SD) of each ROI were recorded, and the mean values of the three ROIs were taken as the CT value and SD value of the renal arteries and erector spinae. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of renal artery were calculated with the SD values of the right erector spinae as background image noise. The SNR of renal artery was defined as the ratio of the CT value of renal artery divided by the SD value of renal artery (SNR = CT renal artery/SD renal artery), while the CNR of renal artery was defined as the difference of CT value between renal artery and erector spinae divided by the SD value of erector spinae (CNR = [CT renal artery ¡ CT erector spinae]/SD erector spinae). Subjective Image Evaluation For the subjective evaluation of image quality, the volume rendering and maximum intensity projection images were generated on the AW4.6 workstation, in addition to the original axial images. The image quality was scored subjectively and blindly on the axial images, volume rendering images and maximum intensity projection images in terms of vessel contrast, image artifacts, image noise, and diagnostic confidence through a comprehensive single subjective score by two radiologists with more than 10 years of working experience using a five-point scoring criterion. The
ASIR-V AND 80KV IN RENAL CTA
scoring criteria are listed in Table 1. In the five-point scoring scale, three points and more were considered to have clinically diagnostic image quality, two points and below did not meet the requirement for clinical diagnosis. Radiation Dose and Contrast Agent Estimation
To estimate the radiation dose, the volumetric CT dose index (CTDIvol) and dose length product (DLP) of the two groups were recorded and the effective dose (ED) was calculated using the formula ED = DLP £ k, of which k represents the radiation dose conversion factor with a value of 0.015 mSv/(mGy.cm) for the abdomen (18). The patient weight in the two groups was recorded, and the contrast agent dose was calculated according to the weight using the formula: contrast agent (g) = contrast agent dose (mgI/kg) £ weight (kg) £ 0.001. Statistical Analysis
All the measurements were analyzed by using SPSS22.0 statistical analysis software. Continuous data were expressed in the form of mean § SD. The difference of the patient characteristics (except for sex which was tested using x2 test), radiation dose and contrast agent between group A and group B were compared using the independent samples t test. The CT values, SD values, SNR values and CNR values of all reconstructed images was analyzed by one-way ANOVA and LSD-t test was used to compare every two reconstruction pairs. The subjective scores for all reconstructed images in terms of vessel contrast, artifacts, noise, and diagnostic confidence by two radiologists were compared with Kruskal-Wallis H test, and the consistency of subjective score between the two radiologists was analyzed using the kappa test, with kappa value 0.75 indicating an excellent consistency, 0.4< kappa value <0.75 indicating good consistency, and kappa value 0.4 indicating poor consistency. For all statistical tests, p < 0.05 being statistically significant. RESULTS Patient Characteristics, Radiation Dose, and Contrast Agent
The patient characteristics, radiation dose, and contrast agent of two groups were listed in Table 2. The patient characteristics including sex, age, weight, height, and BMI showed no statistically significant difference between the two groups
TABLE 1. The Subjective Score Criteria of Image Quality Scores
Image Quality
Vessel Contrast
Artifact
Noise
Diagnostic Confidence
5 4 3 2 1
Best Good Moderate Poor Worse
Very clear Clear Less clear Unclear Cannot be displayed
Very little Mild Moderate Obvious Severe
Very little Mild Moderate Obvious Severe
Fully diagnostic Good diagnostic Diagnostic Affecting diagnosis Nondiagnostic
3
ARTICLE IN PRESS Academic Radiology, Vol &, No &&, && 2019
REN ET AL
TABLE 2. The Comparison of Patients Characteristics, Radiation Dose and Contrast Agent Between Two Groups Parameters Patients characteristics Sex (Male/Female) Age (Year) Weight (kg) Height (cm) BMI (kg/㎡) Radiation dose CTDIvol (mGy) DLP (mGycm) ED (mSv) Contrast agent (g)
Group A
Group B
x2 value/t Value
p Value
26/14 58.80 § 10.56 63.25 § 6.27 168.90 § 6.57 22.24 § 2.66
21/19 56.78 § 16.73 62.63 § 8.48 166.75 § 4.60 22.52 § 2.92
1.289a 0.647b 0.375b 1.695b -0.438b
0.256 0.519 0.709 0.094 0.663
13.00 § 4.08 402.46 § 126.28 6.04 § 1.89 37.95 § 3.76
5.33 § 0.29 140.88 § 12.99 2.11 § 0.19 21.92 § 2.97
11.855b 13.032b 13.032b 21.162b
<0.001# <0.001# <0.001# <0.001#
a x2 value. b t value. # p < 0.05 with independent samples t test.
(p > 0.05). There were significant differences in CTDIvol, DLP, ED and contrast agent between group A and group B (all p < 0.001), the CTDIvol, DLP, ED in group B were 59.0%, 65.0%, and 65.1% lower than those in group A, respectively (all p < 0.001),and the contrast agent dose in group B was 42.2% lower than that in group A (p < 0.001). The relationship between CTDI and BMI in group A and group B were showed in Figure 3 ab.
Objective Image Evaluation
The comparison of CT values, SD values, SNR values, and CNR values between two groups with different reconstruction were listed in Table 3. There was no significant difference in CT values of the left renal artery and right renal artery between the two groups for all reconstructed images (p > 0.05), while the difference in SD values, SNR values, and CNR values between two groups with different reconstruction algorithms were statistically significant (p < 0.05). Compared with FBP reconstruction in group A, the SD value in group B with FBP reconstruction was significantly higher (p < 0.05), SNR value was slightly lower (p > 0.05) while CNR value was significantly lower (p < 0.05). With the increase of ASIR-V percentage in group B, CT values of renal artery showed no significant difference, while the SD values gradually decreased, and SNR values and CNR values gradually increased. The SD values in group B with 40% ASIR-V to 100% ASIR-V reconstructions were statistically lower than that in group A (p < 0.05), SNR values with 50% ASIR-V to 100% ASIR-V reconstruction and CNR values with 70% ASIR-V to 100% ASIR-V reconstruction were significantly higher than those in group A (p < 0.05). Subjective Image Evaluation
The subjective scores of renal CTA are listed in Table 4, and subjective scores of all the reconstructed images with FBP algorithm and ASIR-V reconstruction were more than three points. The 4
subjective scores of reconstructed images in terms of vessel contrast, artifact, noise, and diagnostic confidence were well consistent between the two radiologists (kappa value >0.75, p < 0.05). There were significant differences in the subjective scores of in terms of vessel contrast, artifacts, noise, and diagnostic confidence between the two groups with different reconstruction algorithms (p < 0.05). Compared with FBP reconstruction in group A, the scores of artifacts and noise and diagnostic confidence with FBP reconstruction in group B were significantly lower (p < 0.05), while the score of vessel contrast with FBP reconstruction in group B was slightly lower (p > 0.05). The subjective scores of renal CTA in group B with 60% ASIR-V to 90% 0X3D X ASIR-V reconstruction in terms of vessel contrast, artifacts, noise, and diagnostic confidence were significantly higher than those of FBP reconstruction in group A (p < 0.05), of which the 70%ASIR-V reconstruction achieved the highest comprehensive subjective scores (Fig. 1 and 2).
DISCUSSION RenalCTA, as a fast and noninvasive imaging method, has the characteristic of excellent spatial resolution and high density resolution, which brings the clear vessel contrast between renal artery and its surrounding tissue and help to display the anatomy and abnormality of renal artery clearly. Nevertheless, the main disadvantages of CTA are ionizing radiation exposure and potential contrast-induced nephropathy (19). Therefore, it is very important to reduce radiation dose and contrast agent dose in the process of renal CTA while meeting clinical diagnostic image quality. FBP, as a traditional and standard method for CT image reconstruction, is difficult to find a balance between high image quality and low radiation dose. IR is a new CT reconstruction algorithm, which has greater potential in reducing image noise, suppressing artifacts, and reducing radiation dose compared with traditional FBP algorithm. ASIR, being one of the early IR algorithms that based on noise model, has been widely used in clinical work, which not only helps to reduce radiation dose
Reconstruction Algorithm
CT Value (HU)
Noise (SD Value)
Signal to Noise Ratio (SNR)
Contrast to Noise Ratio (CNR)
Right Renal Artery Left Renal Artery Right Renal Artery Left Renal Artery Right Renal Artery Left Renal Artery Right Renal Artery Left Renal Artery 271.78 § 66.55 282.79 § 54.25 282.79 § 54.25 282.04 § 54.29 283.03 § 53.08 283.46 § 53.05 283.30 § 53.14 282.72 § 53.22 282.43 § 53.37 281.88 § 53.51 281.40 § 53.68 280.16 § 51.42 0.138 0.986
273.32 § 65.28 289.94 § 49.81 289.53 § 49.82 289.11 § 49.95 288.29 § 50.67 290.01 § 49.57 289.90 § 49.97 288.75 § 51.33 289.43 § 50.51 289.36 § 50.60 289.06 § 50.80 288.76 § 51.31 0.320 0.982
23.07 § 3.07 29.78 § 6.42* 27.41 § 6.09* 24.81 § 5.38 22.82 § 5.10 20.49 § 4.15* 18.38 § 3.97* 16.15 § 3.69* 14.04 § 3.58* 12.01 § 3.47* 10.23 § 3.48* 8.62 § 3.14* 96.206 <0.001|
23.13 § 3.63 29.93 § 4.48* 27.71 § 4.22* 25.08 § 3.89* 23.12 § 3.87 20.74 § 3.06* 18.63 § 3.08* 16.27 § 3.19* 14.21 § 3.40* 12.04 § 3.73* 10.34 § 4.04* 8.41 § 4.57* 135.155 |
11.99 § 3.38 9.93 § 2.75 10.79 § 2.88 11.85 § 3.20 12.94 § 3.21 14.28 § 3.47 15.95 § 3.80* 18.15 § 4.27* 21.02 § 5.13* 24.96 § 6.94* 30.47 § 11.23* 37.11 § 18.10* 56.922 <0.001|
12.07 § 3.30 9.81 § 1.72 10.59 § 1.85 11.69 § 2.08 12.67 § 2.28 14.16 § 2.45 15.80 § 2.75* 18.14 § 3.41* 21.04 § 4.25* 25.52 § 6.62* 31.00 § 10.42* 42.27 § 20.18* 73.636 <0.001|
12.26 § 4.44 7.40 § 2.03* 7.97 § 2.22* 8.81 § 2.53* 9.60 § 2.73* 10.75 § 3.17 11.96 § 3.69 13.67 § 4.43 15.79 § 5.51* 18.48 § 7.46* 21.83 § 10.36* 25.48 § 11.33* 38.464 <0.001|
12.34 § 4.32 7.62 § 1.90* 8.21 § 2.07* 9.09 § 2.35* 9.82 § 2.61* 11.05 § 2.97 12.28 § 3.46 13.99 § 4.19 16.22 § 5.17* 19.01 § 7.06* 22.45 § 9.87* 26.25 § 11.08* 44.563 <0.001|
* p < 0.05 with FBP algorithm and ASIR-V algorithm in group B compared with FBP algorithm in group A. | p < 0.05 with one-way ANOVA.
TABLE 4. The Comparison of Subjective Scores of Renal CTA With Different Reconstruction Algorithms Reconstruction Algorithm
Artifact
Noise
Diagnostic Confidence
Radiologist 1
Radiologist 2
Radiologist 1
Radiologist 2
Radiologist 1
Radiologist 2
Radiologist 1
Radiologist 2
4.05 § 0.78 3.95 § 0.75 3.98 § 0.77 4.00 § 0.82 4.03 § 0.70 4.18 § 0.71 4.40 § 0.59* 4.78 § 0.42* 4.93 § 0.27* 4.73 § 0.45* 4.45 § 0.55* 4.28 § 0.45 69.064 <0.001❀
4.10 § 0.74 3.90 § 0.78 3.93 § 0.80 4.03 § 0.79 4.08 § 0.80 4.23 § 0.62 4.43 § 0.64* 4.70 § 0.46* 4.90 § 0.30* 4.88 § 0.34* 4.50 § 0.51* 4.33 § 0.47 69.199 <0.001❀
3.90 § 0.69 3.40 § 0.49* 3.52 § 0.50* 3.60 § 0.52* 3.93 § 0.57 4.10 § 0.55 4.28 § 0.51* 4.50 § 0.56* 4.80 § 0.41* 4.69 § 0.46* 4.63 § 0.49* 4.33 § 0.47* 70.640 <0.001❀
3.93 § 0.73 3.45 § 0.50* 3.58 § 0.51* 3.73 § 0.45 4.03 § 0.66 4.13 § 0.52 4.38 § 0.54* 4.63 § 0.54* 4.78 § 0.42* 4.73 § 0.45* 4.55 § 0.50* 4.35 § 0.48* 64.740 <0.001❀
3.95 § 0.69 3.38 § 0.49* 3.55 § 0.50* 3.88 § 0.61* 3.88 § 0.61 4.12 § 0.56 4.25 § 0.50* 4.43 § 0.55* 4.70 § 0.46* 4.75 § 0.44* 4.85 § 0.36* 4.93 § 0.27* 106.941 <0.001❀
3.92 § 0.70 3.45 § 0.51* 3.53 § 0.51* 3.68 § 0.47* 3.98 § 0.70 4.15 § 0.51 4.35 § 0.53* 4.55 § 0.55* 4.68 § 0.47* 4.83 § 0.39* 4.85 § 0.36* 4.90 § 0.30* 98.650 <0.001❀
4.25 § 0.63 3.68 § 0.47* 3.70 § 0.46* 3.83 § 0.39* 3.88 § 0.46* 4.10 § 0.63 4.40 § 0.49 4.58 § 0.50* 4.89 § 0.30* 4.73 § 0.45* 4.65 § 0.48* 4.34 § 0.48 60.644 <0.001❀
4.30 § 0.56 3.70 § 0.46* 3.78 § 0.42* 3.88 § 0.34* 3.98 § 0.66* 4.18 § 0.55 4.45 § 0.50 4.65 § 0.48* 4.82 § 0.42* 4.73 § 0.45* 4.58 § 0.50* 4.50 § 0.51 40.265 <0.001❀
* p < 0.05 with FBP algorithm and ASIR-V algorithm in group B compared with FBP algorithm in group A. ❀ p < 0.05 with Kruskal-Wallis H test.
ASIR-V AND 80KV IN RENAL CTA
120kV-FBP 80kV-FBP 10%ASIR-V 20%ASIR-V 30%ASIR-V 40%ASIR-V 50%ASIR-V 60%ASIR-V 70%ASIR-V 80%ASIR-V 90%ASIR-V 100%ASIR-V Z value P value
Vessel Contrast
ARTICLE IN PRESS
120kV-FBP 80kV-FBP 10%ASIR-V 20%ASIR-V 30%ASIR-V 40%ASIR-V 50%ASIR-V 60%ASIR-V 70%ASIR-V 80%ASIR-V 90%ASIR-V 100%ASIR-V F value P value
Academic Radiology, Vol &, No &&, && 2019
TABLE 3. The Comparison of Objective Parameters of Renal CTA With Different Reconstruction Algorithms
ARTICLE IN PRESS REN ET AL
Academic Radiology, Vol &, No &&, && 2019
Figure 1. (a) A women aged 59 years with small renal cyst in right kidney, the patient was scanned with 120 kV and 600 mgI/kg and the images was reconstructed with FBP algorithm, volume render (a) showed mild noise and the subjective scores were fourpoint by two radiologists. (be), A 61year-old man with small renal cyst in left kidney, the patient was performed with 80 kV and 350 mgI/kg and the images were reconstructed with FBP (b), 40%ASIR-V (c), 70%ASIR-V(d), 100%ASIR-V (e), the subjective scores for volume render (be) were 3-point, 4-point, 5-point, 4-point by two radiologists, respectively.
and but also helps to provide better image quality. However, ASIR algorithm with higher percentage has tendency to over smooth the image quality, which results in artificial textures or blotchy appearance (20). Model-based iterative reconstruction (MBIR) is another advanced iterative reconstruction algorithm, which relies on more complex and accurate models than ASIR algorithm (21), and which has greater potential to reduce image noise and improve spatial resolution than ASIR algorithm. However, MBIR algorithm requires much longer time for image reconstruction (22). In recent years, an improved ASIR-V, as a new generation IR algorithm, has been developed to balance the model complexness and reconstruction time. ASIR-V algorithm discards optical model and adopts more advanced noise model with the addition of object modeling and physics modeling to balance image noise and spatial resolution when compared with ASIR, which provides better image quality and lower radiation dose than ASIR reconstruction and shorter reconstruction time than MBIR (23). In this study, reducing the tube voltage from 120 kV to 80 kV realized the reduction of radiation dose by 65.1% and contrast agent dose reduction by 42.2%, which was related to that the radiation dose was directly proportional to the square of tube voltage (24), that is to say, slight change in tube voltage would cause large change in radiation dose. The lower tube voltage was closer to the k edge of iodine (33.2 keV) (25), increasing of photoelectric effect and the reducing Compton scattering, and increasing the X-ray attenuation of iodine, which improves the CT values of renal CTA and the 6
vessel contrast between renal artery and surrounding tissues. The increased attenuation of iodine was used to reduce the contrast agent dose in renal CTA in our study. Our results were similar with that of Nagayama Y (26) who reported that the radiation dose decreased by 33% and 65%, respectively when reducing tube voltage from 120 kV to 100 kV or 80 kV, and that of Thor D (27) who reported contrast agent decrease by 40%-44% by using 80 kV. All these studies indicated that the radiation dose and contrast agent could be significantly reduced by using lower tube voltage. The reduction of radiation dose in our study was much greater than that of Nakaura T (28) who reported that the radiation dose was only reduced by 20% via reducing tube voltage from 120 kV to 80 kV, and that of Del Gaizo A J (29) who discovered that using 80 kV could achieve a 30% reduction of radiation dose. The reduction of contrast dose in our study was also much greater than that of Feng C (30) who confirmed that contrast agent was reduced by 26.1% using 80 kV or 100 kV according to different BMI, all the researches performed by Nakaura T, Del Gaizo A J, and Feng C about reducing radiation dose and contrast agent was lower than the results of our study, which may be related to the differences of patient population in different studies, different types or injection rates of contrast agent, and differences caused by different scanners or scanning protocols. For comparison of FBP reconstruction between 120 kV and 80 kV, the results showed that SD values of 80 kV with FBP reconstruction was slightly higher than that of 120 kV, while SNR values and CNR values of 80 kV with FBP
ARTICLE IN PRESS Academic Radiology, Vol &, No &&, && 2019
Figure 2. (a) 57yearold man with 120 kV and FBP algorithm, the subjective scores of maximum intensity projection were fourpoint by two radiologists. (be), 56yearold woman with 80 kV, the images were reconstructed with FBP (b), 40%ASIR-V (c), 70%ASIR-V (D), 100%ASIR-V (e), the subjective scores of maximum intensity projection (be) were 3-point, 4-point, 5-point, 4-point by two radiologists, respectively.
reconstruction was lower than those of 120 kV. In addition, the subjective scores of 80 kV with FBP reconstruction in vessel contrast, artifact, noise, and diagnostic confidence were
ASIR-V AND 80KV IN RENAL CTA
significantly lower than those of 120 kV, which indicated that the reduction of radiation dose by using lower tube voltage would lead to the increase of image noise and the decrease of image quality while maintaining the consistency of other scanning parameters. With the increase of ASIR-V reconstruction percentage in group B, the SD values decreased gradually, and the SNR values and CNR values increased gradually within a certain range from 10%ASIR-V to 70%ASIR-V, indicating that ASIR-V reconstruction could significantly reduce image noise and improve image quality. Compared to the FBP reconstructions in group A with 120 kVp, the SD values with 70% ASIR-V to 100% ASIR-V reconstruction in group B was lower, SNR values and CNR values were significantly higher, which indicated that choosing higher percentage of ASIR-V reconstruction in 80 kV renal CTA could achieve better image quality than FBP reconstructions in 120 kVp. With the increase of ASIR-V percentage in group B, the subjective score of image noise increased gradually from 10%ASIR-V to 100%ASIR-V reconstruction, while the subjective score in vessel contrast, artifacts ,and diagnostic confidence gradually increased from 10%ASIR-V to 70%ASIR-V reconstruction, and then decreased gradually when the ASIR-V reconstruction percentage exceeding 70%. Namely, the 70%ASIR-V reconstruction algorithm could achieve the highest subjective score of overall image quality in 80 kV renal CTA when compared to FBP reconstruction in 120 kV and other ASIR-V reconstruction percentage in 80 kV. Therefore, 70%ASIR-V reconstruction algorithm could achieve the best renal CTA image quality combining subjective evaluation with objective evaluation. Our results showed that lower percentage ASIR-V reconstruction would limit the ability of suppressing image noise and higher percentage ASIR-V reconstruction would result in waxy artifacts or blocky appearance to affect the vessel contrast. However, the research carried by Benz D C (31) reported that 100%ASIR-V reconstruction produced
Figure 3. (ab) The Relationship of CTDI and patient BMI in group A (a) and group B (b).
7
ARTICLE IN PRESS Academic Radiology, Vol &, No &&, && 2019
REN ET AL
substantial noise reduction and improved image quality in lowdose coronary computed tomography angiography, while the research performed by Pontone G (32) discovered that 60%ASIR-V reconstruction provided the optimal balance between image objective parameters and image quality. The results of our study demonstrated the inconsistency with the researches performed by Benz D C and Pontone G, which could be explained by that our study focused on renal CTA and the research by Benz and Pontone focused on coronary artery, and the contrast agent concentration and injection rate would also affect the results of both objective and subjective image quality evaluation (33). This study has some limitations. Firstly, the sample size of our study was small; secondly, the patients population were randomly divided into two groups, which was not paired according to individual differences; thirdly, only normal renal CTA were included into the study population and renal artery disease was not assessed; finally, all the patients evaluated in our study had with small BMI values and patients with larger BMI values were not concluded in the study due to the population characteristics in China, this might introduce bias in our study. Therefore, in the future research work, we need to further expand the sample size to include patients with large BMI values, and the patient population will be paired to divide into groups according to the height, weight and age to reduce the impact of individual differences on the results of the study, and the researches about renal CTA using 80 kV will recruit patients with renal artery diseases. In conclusion, using 80 kV tube voltage with combination of ASIR-V algorithm can significantly reduce radiation dose and contrast agent dose as well as improve image quality in renal CTA for thin patients when compared to FBP using 120 kV. Compared to FBP using 120 kV or 80 kV and other ASIR-V reconstruction percentage, the 70% ASIR-V reconstruction percentage was the best reconstruction algorithm in 80 kV renal CTA for slim patients to achieve the optimal balance between radiation dose and image quality, which helps to provide support for further reduction of radiation dose and contrast agent dose in renal CTA as well as cost saving, and has potential to reduce clinical radiation exposure injury and contrastinduced nephropathy. ACKNOWLEDGMENTS The authors wish to thank Dr. Gu Jun for guidance and suggestions in this paper and Dr. Li Jianying for the modifying and editing the article. REFERENCES 1. 2.
3. 4.
8
Xu Y, Yu S, Zhang L, et al. Application value of iterative reconstruction with CTA to intractable headache patients. Exp Ther Med 2018; 16:603–608. Marenzi G, Assanelli E, Campodonico J, et al. Contrast volume during primary percutaneous coronary intervention and subsequent contrastinduced nephropathy and mortality. Ann Intern Med 2009; 150:170–177. Stacul, Fulvio, Claudio, et al., et al. Subclinical acute kidney injury (AKI) due to iodine-based contrast media. Eur Radiol. 2013; 23:319–323. Piechowiak EI, Peter JF, Kleb B, et al. Intravenous iodinated contrast agents amplify DNA radiation damage at CT. Radiology 2015; 275:692–697.
5.
6.
7. 8.
9.
10.
11. 12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
Raman SP, Mahesh M, Blasko RV, et al. CT scan parameters and radiation dose: practical advice for radiologists. J Am Coll Radiol 2013; 10:840–846. Nijhof WH, Baltussen EJ, Kant IM, et al. Low-dose CT angiography of the abdominal aorta and reduced contrast medium volume: assessment of image quality and radiation dose. Clin Radiol 2016; 71:64–73. Mayosmith WW, Hara AK, Mahesh M, et al. How I do it: managing radiation dose in CT. Radiology 2014; 273:657–672. Yan C, Xu J, Liang C, et al. Radiation Dose Reduction by Using CT with iterative model reconstruction in patients with pulmonary invasive fungal infection. Radiology 2018; 288:285–292. Iyama Y, Nakaura T, Yokoyama K, et al. Low-contrast and low-radiation dose protocol in cardiac computed tomography: usefulness of low tube voltage and knowledge-based iterative model reconstruction algorithm. J Comput Assist Tomogr 2016; 40:941–947. Chang KJ, Collins S, Li B, et al. Optimizing CT technique to reduce radiation dose: effect of changes in kVp, iterative reconstruction, and noise index on dose and noise in a human cadaver. Radiol Phys Technol 2017; 10:180–188. Geyer LL, Schoepf UJ, Meinel FG, et al. State of the art: iterative CT reconstruction techniques. Radiology 2015; 276:339–357. Patino M, Fuentes JM, Singh S, et al. Iterative reconstruction techniques in abdominopelvic CT: technical concepts and clinical implementation. Am J Roentgenol 2015; 205:W19–W31. Mitsumori LM, Shuman WP, Busey JM, et al. Adaptive statistical iterative reconstruction versus filtered back projection in the same patient: 64 channel liver CT image quality and patient radiation dose. Eur Radiol 2012; 22:138–143. Pickhardt PJ, Lubner MG, Kim DH, et al. Abdominal CT with modelbased iterative reconstruction (MBIR):initial results of a prospective trial comparing ultralow-dose with standard-dose imaging. Am J Roentgenol 2012; 199:1266–1274. Qi LP, Li Y, Tang L, et al. Evaluation of dose reduction and image quality in chest CT using adaptive statistical iterative reconstruction with the same group of patients. Br J Radiol 2012; 85:906–911. Marco PD, Origgi D. New adaptive statistical iterative reconstruction ASiRV: assessment of noise performance in comparison to ASiR[J]. J Appl Clin Med Phys 2018; 19:275–286. Kwon H, Cho J, Oh J, et al. The adaptive statistical iterative reconstruction-V technique for radiation dose reduction in abdominal CT: comparison with the adaptive statistical iterative reconstruction technique. Br J Radiol 2015; 88:20150463. Saman R, Co SJ, Simon B. Reduced kilovoltage in computed tomography-guided intervention in a community hospital: effect on the radiation dose. Can Assoc Radiol J 2014; 65:345–351. Pang L, Zhao Y, Dong H, et al. High-Pitch Dual-Source Computed Tomography Renal Angiography Comparison With Conventional Low-Pitch Computed Tomography Angiography: Image Quality, Contrast Medium Volume, and Radiation Dose. J Comput Assist Tomogr 2015; 39:737–740. Kim HG, Yong EC, Lee YH, et al. Quantitative analysis of the effect of iterative reconstruction using a phantom: determining the appropriate blending percentage. Yonsei Med J 2015; 56:253–261. Kim JH, Choo KS, Moon TY, et al. Comparison of the image qualities of filtered back-projection, adaptive statistical iterative reconstruction, and model-based iterative reconstruction for CT venography at 80 kVp. Eur Radiol 2016; 26:2055–2063. Gatti M, Marchisio F, Fronda M, et al. Adaptive statistical iterative reconstruction-V versus adaptive statistical iterative reconstruction: impact on dose reduction and image quality in body computed tomography. J Comput Assist Tomogr 2018; 42:191–196. Lim K, Kwon H, Cho J, et al. Initial phantom study comparing image quality in computed tomography using adaptive statistical iterative reconstruction and new adaptive statistical iterative reconstruction v. J Comput Assist Tomogr 2015; 39:443. Ippolito D, Talei FC, Fior D, et al. Low kV settings CT angiography (CTA) with low dose contrast medium volume protocol in the assessment of thoracic and abdominal aorta disease: a feasibility study. Br J Radiol 2015; 88:20140140. Hou P, Feng X, Liu J, et al. Low tube voltage and iterative model reconstruction in follow-up CT angiography after thoracic endovascular aortic repair: ultra-low radiation exposure and contrast medium dose. Acad Radiol 2018; 25:494–501. Nagayama Y, Oda S, Nakaura T. Radiation dose reduction at pediatric CT: use of low tube voltage and iterative reconstruction. Radio graphics 2018; 38:1421–1440.
ARTICLE IN PRESS Academic Radiology, Vol &, No &&, && 2019
Thor D, Brismar TB, Fischer MA. Low tube voltage dual source computed tomography to reduce contrast media doses in adult abdomen examinations: a phantom study. Med Phys 2015; 42:5100– 5109. 28. Takeshi N, Kazuo A, Natsuki M, et al. Abdominal dynamic CT in patients with renal dysfunction: contrast agent dose reduction with low tube voltage and high tube current-time product settings at 256-detector row CT. Radiology 2011; 261:467–476. 29. Del Gaizo AJ, Fletcher JG, Yu L, et al. Reducing radiation dose in CT enterography. Radiographics 2013; 33:1109–1124. 30. Feng C, Zhu D, Zou X, et al. The combination of a reduction in contrast agent dose with low tube voltage and an adaptive statistical iterative
ASIR-V AND 80KV IN RENAL CTA
27.
31.
32.
33.
reconstruction algorithm in CT enterography: effects on image quality and radiation dose. Medicine 2018; 97:e0151. €ni C, Mikulicic F, et al. Adaptive statistical iterative reconBenz DC, Gra struction-V: impact on image quality in ultralow-dose coronary computed tomography angiography. J Comput Assist Tomogr 2016; 40:958–963. Pontone G, Muscogiuri G, Andreini D, et al. Impact of a new adaptive statistical iterative reconstruction (ASIR)-V algorithm on image quality in coronary computed tomography angiography. Acad Radiol 2018. pii: S1076-6332 (18)30094-1. Schwarz F, Grandl K, Arnoldi A, et al. Lowering radiation exposure in CT angiography using automated tube potential selection and optimized iodine delivery rate. AJR Am J Roentgenol 2013; 200:W628–W634.
9