Abstract
Objectives:
To evaluate the high-helical pitch, multi-energy (ME) scanning mode of a clinical dual-source photon-counting-detector (PCD) CT and the benefit for virtual monoenergetic images (VMIs) for low-contrast dose coronary CTA.
Materials & Methods:
High pitch (3.2) ME coronary CTA was performed in PCD CT in 27 patients using low contrast dose (30 mL of iohexol 350 mg/ml) and 26 patients at routine contrast dose (60 mL). Low-energy-threshold 120 kV images (also known as T3D images), and 50 and 100 keV VMIs were reconstructed using a 1024×1024 matrix and 0.6 mm slices. CT numbers, noise, and contrast to noise ratio (CNR) was measured in ascending aorta (AA), left main coronary artery (LMCA), and distal left anterior descending (LAD) artery. Confidence in grading luminal stenosis with calcific plaque, non-calcific plaque and stent was evaluated by two independent readers on a 0–100 scale (0 the lowest), and a CAD-RADS score assigned. Image contrast enhancement, sharpness, noise, artifacts, and overall image quality were rated using a 5-point ordinal scale (1 the lowest).
Results:
The radiation dose (CTDI) in low and routine-contrast cohorts were 2.5±0.6 mGy and 3.1±1.7 mGy respectively (p= 0.12). At all measured locations, the mean CT number was > 300 HU in 120 kV (LMCA 382.9±76.2, Distal-LAD 341.0±53.9, AA-399.5±76.1) and 50 keV images (LMCA 667.5±139.9, distal LAD 578.1±121.5, AA 700.8±142.5) in the low contrast cohort, with a 96 % increase in CT numbers for 50 keV over 120 kV. CT numbers were significantly higher (p < 0.0001) in 50 keV than 120 kV and 100 keV VMI. CNR was also significantly (p<0.0001) higher in 50 keV than 120 kV and 100 keV images in all vessels. Confidence in assessment of luminal stenosis in the presence of calcific plaque was significantly higher (p=0.001) with the addition of 100 keV VMI (median score 100) than using 50 keV alone (median score 70) and 120 kV (median score 70) for reader 1, but no significant differences were seen for Reader 2 who had similar median scores of 100 for all image types. The confidence in assessment of luminal stenosis within a stent improved with the use of 100 keV images for both readers (Reader 1: median scores for 50+100 keV = 100; 50 keV= 82.5; 120 kV = 82.5; Reader 2: 50+100 keV = 100, 50 keV = 90; 120 kV = 90). There were no significant differences in confidence scores for assessment of luminal stenosis from non-calcific plaques for both the readers. The reader-averaged qualitative scores for vascular enhancement and overall image quality were significantly higher for 50 keV VMI than 120 kV images in both low and routine contrast dose cohorts. The image sharpness was non-significantly higher at 50 keV VMI than 120 kV images and the artifact score was comparable for 50 keV VMI and 120 kV images. The noise was higher in 50 keV VMI than 120 kV images.
Conclusions:
High-pitch ME PCD-CT mode produced diagnostic quality coronary CTA images at low radiation and iodinated contrast doses. Availability of ME VMI images significantly improved CNR, overall image quality, and confidence in assessment of luminal stenosis in the presence of calcific plaques and stents and resulted in change of CAD-RADS categories in 9 patients.
Keywords: Coronary CTA, atherosclerosis, iodine, virtual monoenergetic image, photon counting detector CT, calcium blooming
INTRODUCTION
Coronary CT angiography (CCTA) is a well-established technique in the evaluation of coronary artery disease (CAD), especially in the low-to-intermediate risk population (1,2). Data from clinical trials and multi-society guidelines have cemented the role of CCTA as a first-line imaging test in the evaluation of chest pain (1,2). Although recent studies have shown that the risk of acute kidney injury following iodinated contrast media is negligible, it is still considered a relative contraindication in patients with severe renal dysfunction (3). Hence, there is a need for low-contrast-dose strategies to minimize the risk of renal injury, especially with the high prevalence of comorbidities such as diabetes and hypertension in patients referred for CCTA. The recent issues with contrast shortage also highlight the need for low-contrast-dose protocols (4). Several strategies have been evaluated for decreasing contrast dose including low tube voltage, high-pitch helical mode, individualized tube voltage, and low-energy virtual monoenergetic images (VMIs) from a dual-energy CT scanner (5–9).
The photon-counting detector (PCD) CT system is a novel technology that uses semiconductor detectors to directly convert x-ray photons to electrical signals without the generation of visible light (10). Using energy-specific thresholds, each detected photon is allocated to a specific energy bin, allowing the acquisition of multi-energy information. PCD-CT allows dose-efficient high-spatial resolution images due to absence of septa between individual detector elements (11, 12). The high-resolution mode has shown promising results in the evaluation of coronary arteries with heavily calcified plaques and stents as well as characterization of plaques (13–16). The multi-energy mode of PCD CT has shown good results in the evaluation of extracellular volume and peri coronary inflammation (17,18). The first commercial dual-source (DS) PCD CT (NAETOM Alpha, Siemens) has a novel high-pitch helical multi-energy scan mode (Quantumplus), which enables a helical pitch up to 3.2 simultaneous to a multi-energy acquisition at a temporal resolution of 66 milliseconds. In cardiac phantoms, this mode performs material decomposition with excellent geometric and quantitative accuracy, and motion robustness, due to improved temporal resolution and spatiotemporal registration (19). A dynamic phantom study showed the possibility of 50% contrast reduction using low-energy VMI in this mode (20). A thoracic CTA study using this mode showed significantly increased CNR of low-energy VMI compared to conventional energy integrating detector (EID) CT at the same radiation dose (21).
The utility of this high-pitch helical ME mode has not been evaluated yet in clinical coronary CTAs at low contrast dosage. The purpose of this study is to evaluate the combined high-pitch multi-energy mode of the DS-PCD-CT for low-contrast dose coronary CTA. We hypothesized that the availability of VMI will enhance vascular contrast and improve the assessment of coronary artery plaques
MATERIALS & METHODS
This single-center prospective HIPAA-compliant study was approved by our institutional review board; written informed consent was obtained. Subjects were recruited from patients with CAD receiving a clinical CCTA exam using a commercial EID-CT scanner (SOMATOM Force, Siemens Healthcare) between April and December 2021.
CT Acquisition
Research scans were performed on an FDA-cleared dual-source PCD-CT system (NAEOTOM Alpha, Siemens Healthcare, Forchheim, Germany). The primary PCD array has a 50-cm field of view (FOV) and the secondary PCD array has a 36-cm FOV. A heart rate < 70 bpm is required for the high-pitch (3.2) helical scanning (Flash mode). No medications, including beta-blockers and nitroglycerine, were given to these patients. Two different contrast doses were utilized in this study: A) Low dosage of 30 mL of iohexol (Omnipaque 350, iodine concentration 350 mg/mL, GE Healthcare, Madison, USA) at 5 mL/sec followed by 15 mL of saline at 5 mL/sec; B) Routine dosage of 60 mL of iohexol at 5 mL/sec followed by 30 mL of saline at 5 mL/sec. Bolus tracking (CareBolus, Siemens Healthcare) was used to trigger CTA acquisition using a circular region-of-interest (ROI) placed in the ascending aorta, with a CT number threshold of 150 HU at 120 kV. Prospective ECG-triggering in Flash mode was performed with a starting phase of 55 % R-R interval, targeting diastolic phase. Scan parameters were 120 kV, 0.25-second rotation, 3.2 helical pitch, and 144 × 0.4 mm collimation. Automatic exposure control (CAREkeV, Siemens Healthcare GmbH) was turned on, with CAREkeV IQ level of 56, optimized for a vascular exam.
Image Reconstruction
Images were reconstructed from the CT scanner using VA40 software (Siemens Healthcare). The 120 kV low-energy threshold images (called T3D on the scanner) utilize all incident x-ray photons incident on the CT detectors above 20 keV. The use of a lower-energy threshold allows the elimination of electronic noise. These images are analogous to conventional polyenergetic 120 kV images using EID-CT but have higher iodine signal due to equal weighting of all the detected photons in PCD-CT (11,21). To take advantage of the ability of PCD-CT for multi-energy (ME) scanning at high pitch, VMIs at 50 keV and 100 keV were also generated, the former to enhance signal of iodinated contrast and the latter to reduce artifacts from calcium blooming and stents (23,24) (Fig 1). These specific levels were chosen based on our experience in early cases and literature from other multi-energy scanners (23–25). An iterative reconstruction algorithm (QIR) with strength of 4, medium sharp kernel (Bv40), 0.6 mm slice thickness with 0.3 mm increment, FOV of 200 × 200 mm2 and a matrix size of 1024 × 1024 were used. A 1024 × 1024 matrix is unlikely to create noticeable difference to a 512×512 matrix at this specific kernel (13). We used this matrix because this is the standard in our institution.
Figure 1. Types of images evaluated in this paper. 63-year-old woman with chest pain-.
A. Low-energy threshold 120 kV images (T3D) are those which utilize almost all the x-ray photons incident on the CT detectors (20–120 keV photons). B. 50 keV virtual monoenergetic images (VMI) mimic a monoenergetic CT obtained at 50 keV with high signal of iodinated contrast. C. 100 keV VMI mimics monoenergetic images at 100 keV and has lower artifacts, but also has lower iodine signal.
Image Analysis - Quantitative
Quantitative measurements were obtained by a single reader blinded to clinical or qualitative metrics. ROIs were placed in the ascending aorta (AA), left main coronary artery (LMCA), and distal left anterior descending coronary (LAD) artery (Fig 2). The size of the ROI was at-least 2/3rd of the area of the vessel of interest. Enhancement was measured as mean CT number in the ROI and noise was measured as the standard deviation of CT number. Background values were measured in the anterior mediastinal fat adjacent to the ascending aorta. The contrast-to-noise ratio (CNR) was calculated as = .
Figure 2. Quantitative measurements in a 53-year- woman with chronic intermittent chest pain and intermediate probability of coronary artery disease.
Example images showing ROIs placed in the ascending aorta (A), left main coronary artery (LMCA) (B) and distal left anterior descending (LAD) coronary artery (C). The background was measured in the anterior mediastinal fat adjacent to the ascending aorta (A).
Image Analysis - Qualitative
The images were evaluated independently by two fellowship-trained cardiovascular radiologists with 25 and 10 years of experience in different sessions. On a clinical workstation, the 120 kV, 50 keV VMI, and 100 keV VMIs were loaded on separate panels. The radiologist was allowed to adjust and customize the window settings to optimize visualization of the coronary arterial lumen. The confidence in detection and classification of luminal narrowing from non-calcific plaque, calcific plaque, and stent was evaluated with a 0 to 100 scale, with 0 indicating complete absence of confidence and 100 indicating absolute confidence in correctly classifying stenosis. This was evaluated first in the 120 kV image, followed by 50 keV and then by a combination of 50 and 100 keV VMIs. The CAD-RADS scores in 120 kV, 50 keV and 50 keV + 100 keV images were also evaluated (26,27). Contrast enhancement, sharpness, noise, artifacts, and overall image quality were rated on a 5-point ordinal scale, with 1 being the lowest and 5 being the highest (Table 1).
Table 1-.
Subjective image quality analysis of PCD CT images
| Parameter measured | Scores |
|---|---|
| Coronary artery enhancement | 1 = Unacceptable enhancement, nondiagnostic 2 = Poor enhancement, questionably/barely diagnostic 3 = Fair/reasonable enhancement, diagnostic 4 = Good enhancement, fully diagnostic 5 = Excellent, fully diagnostic |
| Coronary artery sharpness | 1 = Severe blur 2 = Moderate blur 3 = Mild blur 4 = Sharp 5 = very sharp |
| Artifacts | 1 = Severe/major, non-diagnostic 2 = Severe/major, diagnosis possible but with low confidence 3 = Moderate, questionably diagnostic 4 = Mild, diagnosis not affected 5= None, fully diagnostic |
| Noise | 1 = Severe/major, non-diagnostic 2 = Severe/major, diagnosis possible but with low confidence 3 = Moderate, questionably diagnostic 4 = Mild, diagnosis not affected 5= None, fully diagnostic |
| Overall image quality | 1 = Unacceptable image quality, nondiagnostic 2 = Poor image quality (motion/noise/low contrast), questionably/barely diagnostic to evaluate lumen 3 = Fair/reasonable image quality (some artifacts are present), but diagnostic to evaluate lumen 4 = good image quality, fully diagnostic to evaluate lumen 5 = Excellent image quality, fully diagnostic |
Statistical Analysis
Statistical analysis was performed using the software R version 4.0.3 (R Core Team, Vienna, Austria). Quantitative data were presented as means and standard deviations or medians and inter-quartile ranges. For quantitative measurements, two-sample unequal variance one-tailed t-test was used to compare routine and low contrast studies and paired one-tailed t-test was used to compare 120 kV, 50 keV VMI, and 100 keV VMI. Inter-reader agreement for presence of calcific plaques and stents was evaluated by absolute agreement and Cohen’s kappa. Intra-rater intermodality agreement for soft plaque and stenosis was evaluated with Cohen’s kappa. Evaluation of differences in confidence scores for classification of luminal stenosis was performed using non-parametric Friedman testing alongside with post-hoc pair-wise comparisons using Wilcoxon signed rank tests. Multiple testing was accounted for using a Holm correction. Inter-reader agreement of quality scores was evaluated using weighted kappa with quadratic weights (with 95 % CIs). P values < 0.05 were considered statistically significant.
RESULTS
Demographics and radiation doses
The study had 27 patients in the low-contrast-dosage cohort (16 males, 11 females) and 26 patients in the routine-contrast-dosage cohort (16 males, 10 females), with mean age of 68.3±9.8 for low-contrast and 64.9±15.6 for routine contrast cohorts. There was no significant difference in BMI between the cohorts (Low-contrast, 27.0 ± 4.5; routine contrast 27.4 ± 4.1, p=0.57). The radiation dose (CTDI) of the low-contrast cohort was 2.5±0.6 mGy and routine-contrast cohort was 3.1±1.7 mGy (p= 0.12) (Table 2).
Table 2.
Demographics and radiation doses of patients in this study
| Low-dose cohort | Routine dose cohort | Entire cohort | |
|---|---|---|---|
| Number | 27 | 26 | 53 |
| Age (Mean ± SD) | 68.3±9.8 | 64.9±15.6 | 66.6±13 |
| Sex – N (%) | |||
| Male | 16 (59) | 16 (62) | 32 (60) |
| Female | 11 (41) | 10 (48) | 21 (40) |
| BMI (kg/m2) | 27.0 ± 4.5 | 27.4 ± 4.1 | 27.3 ± 4.3 |
| Radiation dose, CTDI ± SD (mGy) | 2.5±0.6 | 3.1±1.7 | 2.8±1.2 |
Quantitative Analysis
The mean contrast enhancement was greater than 300 HU in AA, LMCA and distal LAD in 120 kV and 50 keV VMI in both low- and routine-contrast cohorts. The enhancement was significantly higher (p < 0.0001) in 50 keV VMI than 120 kV and 100 keV VMI at all measured locations in both contrast dosage groups (Table 3). There was a 96% mean increase in iodine enhancement in 50 keV VMI over 120 kV in the low-contrast cohort. Noise was also significantly higher (p <0.0001) in 50 keV VMI than 120 kV and 100 keV VMI at all locations in both contrast dosage groups. In low-contrast cohort, the CNR of 50 keV was significantly higher than 120 kV and 100 keV in AA, LMCA, and distal LAD. Similar results were obtained in routine contrast cohort (Table 3).
Table 3-.
Quantitative metrics in PCD-CT images in entire, routine contrast dosage (60mL) and low contrast dosage (30 mL) cohorts.
| CT number | Noise | CNR | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| 120 kV | 50 keV VMI | 100 keV VMI | 120 kV | 50 keV VMI | 100 keV VMI | 120 kV | 50 keV VMI | 100 keV VMI | ||
| LMCA | Low contrast | 382.9 ± 76.2 | 667.5 ± 139.9 | 382.9 ± 76.2 | 23.20 ± 3.19 | 34.83 ± 5.87 | 22.23 ± 3.87 | 19.83 ± 5.75 | 16.97 ± 4.66 | 9.75 ± 2.46 |
| Routine contrast | 449.7 ± 129.2 | 773.2 ± 231.3 | 260.1 ± 73.4 | 24.17 ± 4.01 | 35.41 ± 7.05 | 23.82 ± 4.54 | 23.20 ± 10.41 | 19.68 ± 8.68 | 11.82 ± 5.84 | |
| Combined | 415.0 ± 109.3 | 718.3 ± 194.9 | 234.0 ± 61.4 | 23.67 ± 3.61 | 35.11 ± 6.41 | 22.99 ± 4.24 | 21.45 ± 8.41 | 18.27 ± 6.96 | 10.75 ± 4.49 | |
| Distal LAD | Low contrast | 341.0 ± 53.9 | 578.1 ± 121.5 | 193.2 ± 33.3 | 23.20 ± 3.19 | 34.83 ± 5.87 | 22.23 ± 3.87 | 17.03 ± 4.45 | 15.09 ± 3.73 | 9.01 ± 2.50 |
| Routine contrast | 382.5 ± 93.4 | 677.3 ± 210.4 | 221.2 ± 61.4 | 24.17 ± 4.01 | 35.41 ± 7.05 | 23.82 ± 4.54 | 20.42 ± 9.14 | 16.66 ± 6.58 | 9.99 ± 4.67 | |
| Combined | 361.0 ± 77.6 | 625.8 ± 175.7 | 206.6 ± 50.4 | 23.67 ± 3.61 | 35.11 ± 6.41 | 22.99 ± 4.24 | 18.66 ± 7.24 | 15.85 ± 5.30 | 9.48 ± 3.70 | |
| Ascend-ing aorta | Low contrast | 399.5 ± 76.1 | 700.8 ± 142.5 | 217.4 ± 33.2 | 23.20 ± 3.19 | 34.83 ± 5.87 | 22.23 ± 3.87 | 20.83 ± 5.92 | 17.68 ± 4.69 | 10.10 ± 2.43 |
| Routine contrast | 469.5 ± 132.8 | 822.9 ± 254.3 | 262.4 ± 66.5 | 24.17 ± 4.01 | 35.41 ± 7.05 | 23.82 ± 4.54 | 24.64 ± 11.01 | 20.49 ± 8.79 | 11.86 ± 5.55 | |
| Combined | 433.2 ± 111.8 | 759.5 ± 211.2 | 239.1 ± 56.2 | 23.67 ± 3.61 | 35.11 ± 6.41 | 22.99 ± 4.24 | 22.60 ± 8.95 | 18.95 ± 7.08 | 10.95 ± 4.32 | |
Qualitative Analysis
Confidence in the detection and classification of luminal stenosis
In coronary arteries with calcification and calcium blooming:
For Reader 1, confidence in assessment of luminal stenosis with calcific plaques and calcium blooming in the low-dose cohort was significantly higher (p=0.001) with 50 +100 keV VMIs (median score of 100) than 50 keV alone (median score 70) and 120 kV (median score 70). Similar results were also seen in the routine contrast cohort with 50+100 keV VMI median score of 80, 50 keV score of 50, and 120 kV score of 50 (p=0.0001) (Table 4, Figs, 3, 4)
Table 4-.
Distributional summaries of confidence evaluation of luminal stenosis with calcific plaques
| Low Contrast Group | Routine Contrast Group | Contrast groups combined | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| N | 120 kV | 50 keV | 50+100 keV | N | 120 kV | 50 keV | 50+100 keV | N | 120 kV | 50 keV | 50+100 keV | |
| READER 1 | ||||||||||||
| LMCA | 8 | 70 (70,85) | 70 (70,85) | 95++ (87.5,100) | 12 | 60 (50,80) | 60 (50,80) | 85++ (78.8,100) | 20 | 70* (57.5, 80) | 70 (57.5, 80) | 90++ (80, 100) |
| LAD | 17 | 70 (50,90) | 70 (50,90) | 100++ (70,100) | 17 | 50 (50,70) | 50 (50,70) | 80++ (70,80) | 34 | 70 (50, 77.5) | 70 (50, 77.5) | 80++ (70,100) |
| LCx | 13 | 60 (50,90) | 70 (50,90) | 90++ (70,100) | 15 | 50 (50,70) | 50 (50,70) | 70++ (70,80) | 28 | 50 (50,70) | 55 (50,70) | 80++ (70,100) |
| RCA | 12 | 75 (57.5,90) | 75 (57.5,90) | 95++ (77.5,100) | 14 | 50 (50,70) | 50 (50,70) | 70++ (55,87.5) | 26 | 65 (50,80) | 70 (50,80) | 80++ (70,100) |
| All | 50 | 70 (52.5,90) | 70 (60,90) | 100++ (72.5,100) | 58 | 50 (50,70) | 50 (50,70) | 80++ (70,90) | 108 | 65 (50,80) | 70 (50,80) | 80++ (70,100) |
| READER 2 | ||||||||||||
| LMCA | 9 | 100 | 100 | 100 | 12 | 100 | 100 | 100 | 21 | 100 | 100 | 100 |
| LAD | 16 | 82.5 (75,100) | 90 (75,100) | 100++ (90,100) | 15 | 75 (60,82.5) | 80 (65,87.5) | 80 (77.5,95) | 31 | 80 (75,95) | 80 (75.95) | 90++ (87.5,100) |
| LCx | 11 | 100 (82.5,100) | 100 (90,100) | 100 (95,100) | 14 | 80 (75,97.5) | 87.5 (75,100) | 100++ (90,100) | 25 | 90 (75,100) | 90 (75,100) | 100++ (90,100) |
| RCA | 10 | 100 | 100 | 100 | 14 | 80 (75,100) | 80 (75,100) | 95++ (76.2,100) | 24 | 100 (78.8,100) | 100 (78.8,100) | 100 (88.8,100) |
| All | 46 | 100 (81.2,100) | 100 (90,100) | 100 (96.2,100) | 55 | 80 (75,100) | 90 (75,100) | 100++ (85,100) | 101 | 90 (75,100) | 100 (80,100) | 100 (90,100) |
All values are medians with Q1, Q3 values in parenthesis
indicates this image type was superior to the other two image types.
N indicates number of lesions identified.
Figure 3. Confidence in luminal assessment with calcific lesions.
Charts showing the confidence scores for luminal assessment in calcific lesions for Reader 1 and Reader 2 in low and routine contrast dose cohorts.
Figure 4. Calcium blooming in PCD-CT.
72-year-old- male with chronic intermittent chest pain. Axial oblique reconstructed 120 kV (A, T3D), 50 keV VMI (B) and 100 keV VMI (C) images at the level of LAD show dense calcific plaques which demonstrate decreased calcium blooming and improved luminal delineation with 100 keV VMIs than 50 keV VMI and 120 kV images.
For Reader 2, there was no significant difference in scores between image types in the low contrast dose cohort, with median scores of 100 for 120 kV, 50 keV, and 50 +100 keV. In the routine contrast dose cohort, the scores were higher for 50+100 keV (median score of 100) than 120 kV (median score of 80, p=0.005) and 50 keV (median score of 90, p=0.007) (Fig 3).
In coronary arteries with non-calcific plaque:
For reader 1, the confidence in assessment of luminal stenosis with non-calcific plaque was not different between the different image types in low-contrast cohort with median scores of 100 for all image types. In the routine contrast cohort, the scores were higher for 50 keV and 50+100 keV (median scores of 90 in both) compared to 120 kV (median score of 65).
For reader 2, the confidence in assessment of luminal stenosis with non-calcific plaque was not different between the different image types in low-contrast cohort with scores of 100 for all image types. In the routine contrast cohort, the scores were higher for 50 keV and 50+100 keV (median scores of 87.5 in both) compared to 120 kV (median score of 77.5). For both readers, the addition of 100 keV did not change the scores compared to the scores for 50 keV images only.
In coronary arteries with stents:
Stents were determined to be present in 15 arteries by reader 1 and in 14 arteries by reader 2, with good inter-reader agreement aside for RCA (kappa vaSlues LAD: 0.67, LCx: 0.73, RCA: \0.24).
For Reader 1, improvement in confidence of luminal stenosis assessment within a stent was seen using 50+100 keV VMI compared to 50 keV and 120 kV in low-dose cohort (median scores for 50+100 keV= 100; 50 keV = 82.5; 120 kV = 82.5). There was no difference in scores in routine dose cohort (median score of 70 for all image types).
For Reader 2, there was improvement confidence of luminal stenosis assessment within a stent using 50+100 keV VMIs than 50 keV and 120 kV, both in low-dose (median scores in 50+100 keV = 100, 50 keV = 90; 120 kV = 90) and routine dose (median scores in 50+100 keV = 80; 50 keV = 75; 120 kV = 75). Statistical significance could not be evaluated due to small number of stents in this cohort.
CAD-RADS Score
The CAD-RADS scores for both readers with different image types are shown in Figures 5 and 6.
Figure 5. CAD-RADS score changes between 120 kV (T3D) and 50 keV.
Confusion matrices of CAD-RAD scores at 120 kV (y-axis) and 50 keV (x-axis) images for Reader 1 and Reader 2 at low and routine contrast doses. No change in categories was shown in the low contrast dose for both readers. In routine dose cohort, the score changed from non-diagnostic to diagnostic categories in 1 patient for Reader 1 and 3 patients for Reader 2.
Figure 6. CAD-RADS score changes between 50 keV and 50+100 keV VMIs.
Charts showing the CAD-RAD scores at 50 keV and 50+100 keV VMIs for Reader 1 and Reader 2 at low and routine contrast doses. In low dose cohort, the addition of 100 keV to 50 keV VMI decreased the CAD-RADS score in 3 patients for Reader 1 and 3 patients for Reader 2. In routine dose cohort, addition of 100 keV decreased the CAD-RADS score in 1 patient for Reader 1 and 3 patients for Reader 2 due to decreased calcium blooming.
50 keV vs. 50 +100 keV
For reader 1, in the low contrast cohort, the CAD-RADS score decreased with the addition of 100 keV images in 3 patients (it changed from 4 to 3 in 2 patients and 3 to 2 in one patient); routine contrast cohort, CAD RADS scored decreased from 3 to 2 in one patient.
For reader 2, in the low contrast cohort, the CAD-RADS score decreased with the addition of 100 keV images in 3 patients (one each for change from 2 to 1, 3 to 2, and 4 to 3); in routine contrast, the CAD-RADS score decreased with the addition of 100 keV images in 3 patients (CAD-RADS 4 to 3 in 2 patients and CAD-RADS 3 to 2 in one patient). This decrease in CAD-RADS score was due to decreased calcium blooming in 100 keV images.
50 keV Vs 120 kV
For reader 1, there was no change in CAD-RADS score between 50 keV and 120 kV images in the low contrast cohort, but in the routine contrast cohort one “N” score changed to 0.
For reader 2, there was no change in CAD-RADS score between 50 keV and 120 kV images in the low contrast cohort, but in the routine contrast cohort three “N” scores changed to 0. All N scores were given due to suboptimal enhancement in 120 kV yet were deemed to be of diagnostic quality in 50 keV images (Figure 7).
Figure 7. Improved contrast with 50 keV images.
Axial 120 kV (A, T3D) and 50 keV VMI (B) in a 47-year-old who received 60 mL of iodinated contrast shows suboptimal non diagnostic quality with 120 kV images and a diagnostic quality with 50 keV VMI.
Image Quality Scores
The averaged (both readers) qualitative scores for vascular enhancement and overall image quality were significantly higher for 50 keV VMI than 120 kV images in both low and routine contrast dose cohorts (Table 5). P values for enhancement were < 0.0001 and 0.02 in low and routine contrast cohorts respectively and for overall quality were 0.01 and 0.04 in low and routine contrast cohorts (Table 5). The image sharpness was non-significantly (p=0.16) higher at 50 keV VMI than 120 kV images and the artifact score was comparable for 50 keV VMI and 120 kV images in both low and routine dose cohorts. The noise was higher in 50 keV VMI than 120 kV images in both cohorts. The individual scores of each reader can be seen in Table 5. Note that since the 50 and 100 keV images were viewed together, there is no separate score for 100 keV images. The absolute mean score of 50 keV images in the low contrast cohort was > 4 for all the metrics, indicating the good quality of these images (Table 5, Figure 8).
Table 5-.
Qualitative Scores of 120 kV and 50 keV PCD-CT images for both readers individually and then averaged for the entire cohort, low contrast cohort and routine contrast cohort. Summaries are presented as Mean ± SD.
| Low Contrast Group | Routine Contrast Group | Contrast groups combined | |||||||
|---|---|---|---|---|---|---|---|---|---|
| 120 kV | 50 keV | p-value | 120 kV | 50 keV | p-value | 120 kV | 50 keV | p-value | |
| READER 1 | |||||||||
| Enhancement | 4.4 ± 0.8 | 4.6 ± 0.6 | 0.43 | 4.0 ± 1.1 | 4.1 ± 0.9 | 0.83 | 4.2 ± 1.0 | 4.3 ± 0.8 | 0.50 |
| Sharpness | 4.2 ± 0.8 | 4.2 ± 0.8 | 0.89 | 4.3 ± 0.9 | 4.3 ± 0.9 | 1.00 | 4.2 ± 0.8 | 4.2 ± 0.8 | 0.90 |
| Artifact | 4.0 ± 1.1 | 4.0 ± 1.1 | 0.88 | 3.5 ± 1.3 | 3.5 ± 1.3 | 1.00 | 3.8 ± 1.2 | 3.8 ± 1.2 | 0.90 |
| Noise | 4.7 ± 0.5 | 4.5 ± 0.6 | 0.06 | 4.7 ± 0.6 | 4.4 ± 0.6 | 0.06 | 4.7 ± 0.5 | 4.5 ± 0.6 | 0.01 |
| Overall IQ | 4.1 ± 1.1 | 4.1 ± 1.1 | 0.91 | 3.7 ± 1.4 | 3.7 ± 1.3 | 0.95 | 3.9 ± 1.3 | 3.9 ± 1.2 | 0.99 |
| READER 2 | |||||||||
| Enhancement | 4.0 ± 0.7 | 5.0 ± 0.2 | < 0.0001 | 4.0 ± 1.0 | 4.9 ± 0.4 | < 0.0001 | 4.0 ± 0.9 | 4.9 ± 0.3 | < 0.0001 |
| Sharpness | 4.1 ± 0.8 | 4.5 ± 0.9 | 0.02 | 4.0 ± 0.4 | 4.4 ± 0.6 | 0.02 | 4.1 ± 0.6 | 4.4 ± 0.7 | < 0.0001 |
| Artifact | 4.5 ± 0.8 | 4.6 ± 0.7 | 0.80 | 4.4 ± 0.6 | 4.5 ± 0.6 | 0.80 | 4.5 ± 0.7 | 4.5 ± 0.7 | 0.72 |
| Noise | 4.5 ± 0.6 | 4.6 ± 0.5 | 0.50 | 4.4 ± 0.5 | 4.5 ± 0.5 | 0.60 | 4.4 ± 0.5 | 4.5 ± 0.5 | 0.39 |
| Overall IQ | 3.6 ± 0.6 | 4.7 ± 0.5 | < 0.0001 | 3.6 ± 0.9 | 4.6 ± 0.6 | < 0.0001 | 3.6 ± 0.7 | 4.7 ± 0.6 | < 0.0001 |
| AVERAGED | |||||||||
| Enhancement | 4.2 ± 0.6 | 4.8 ± 0.4 | < 0.0001 | 4.0 ± 0.9 | 4.5 ± 0.5 | 0.02 | 4.1 ± 0.8 | 4.6 ± 0.5 | < 0.0001 |
| Sharpness | 4.1 ± 0.7 | 4.4 ± 0.7 | 0.16 | 4.2 ± 0.5 | 4.3 ± 0.6 | 0.20 | 4.2 ± 0.6 | 4.3 ± 0.7 | 0.06 |
| Artifact | 4.3 ± 0.8 | 4.3 ± 0.8 | 0.80 | 4.0 ± 0.8 | 4.0 ± 0.8 | 0.90 | 4.1 ± 0.8 | 4.2 ± 0.8 | 0.80 |
| Noise | 4.6 ± 0.4 | 4.5 ± 0.4 | 0.51 | 4.6 ± 0.4 | 4.5 ± 0.5 | 0.44 | 4.6 ± 0.4 | 4.5 ± 0.4 | 0.31 |
| Overall IQ | 3.9 ± 0.8 | 4.4 ± 0.7 | 0.01 | 3.7 ± 1.0 | 4.2 ± 0.8 | 0.04 | 3.8 ± 0.9 | 4.3 ± 0.8 | < 0.0001 |
IQ= Image quality
Figure 8. Image quality scores.
A. Stacked barplots showing the distribution of image quality scores (color graded) in 120 kV, 50 keV VMI and 50+100 keV VMI for readers 1 and 2. B. Stacked barplots showing the distribution of image quality scores for readers 1 and 2, stratified by low and routine contrast dose
DISCUSSION
To our knowledge, this is the first study to demonstrate diagnostic quality images in low-contrast dosage coronary CTA utilizing the high-pitch, multi-energy scan mode of a dual-source PCD-CT scanner. 50 keV VMI was superior to 120 kV images with significantly higher vascular enhancement and CNR. 50 keV VMI were able to salvage non-diagnostic contrast-enhanced studies to provide diagnostic CAD-RADS scores. 100 keV VMI improved the diagnostic confidence in evaluation of luminal stenosis associated with calcific plaques and stents and downgraded CAD-RADS scores by reducing blooming artifacts.
Consistent high-quality images utilizing the high-pitch helical ME mode of a PCD-CT scanner was obtained in our study which included patients with heart rate < 70 bpm. ECG-gated high-pitch helical mode has been shown with previous iterations of dual-source CT scanners with EIDs to provide motion-free images of the coronary arteries, with low radiation and iodinated contrast doses compared to other coronary CTA techniques, such as prospective ECG triggering and retrospective ECG gating (28–30). However, the high-pitch mode on EID-CT can only operate in single energy mode as both tubes need to operate at the same kV. With the availability of multi-energy capabilities on PCD-CT, additional image sets can be generated, including full spectrum (120 kV) and virtual monoenergetic images. The ideal type of image, including the specific energy level of VMI, depends on several factors, including the scanner technology, body part, the specific clinical task, and noise (25,31–33). Our study showed that 50 keV images are superior to 120 kV images for overall image quality, vascular enhancement, and coronary artery sharpness with mild increased noise. Knowledge of the preferred image is essential to simplify the workflow, both for the scanning technologist and interpreting cardiac imager.
We have shown a significant improvement in the iodine contrast with 50 keV VMI compared to 120 kV images. It is well established that the contrast increases at low energy images due to increased photoelectric effect and proximity to the K-edge of iodine (34, 35). This allows the use of low dose of iodinated contrast agent, which is beneficial in patients with severe renal dysfunction and helps salvage CT scans that are suboptimally enhanced due to either technical or patient factors. In our study, there were 3 patients who were CAD-RADS score “N” in 120 kV images but received diagnostic CAD-RADS scores with the use of 50 keV VMI. This saves the need for additional doses of iodinated contrast agent or the need for additional scans, which is beneficial for the patient. We did not evaluate images lower than 50 keV VMI due to what we deemed to be higher than acceptable image noise, unnatural noise texture and image appearance in our early experience.
We also demonstrated the added value of 100 keV VMI in evaluating patients with calcium and stent blooming. Blooming artifact is a major challenge in CCTA, leading to overestimation of luminal stenosis in the presence of calcification and stent, which may lead to inappropriate clinical management (23,24). Blooming is primarily due to limits to spatial resolution and the resultant partial volume averaging and can be reduced by several measures including windowing, filtering, or increasing the system spatial resolution. High-energy VMI has also been shown to decrease calcium blooming in CCTA (23,24). We have demonstrated the utility of 100 keV VMIs in improving the diagnostic confidence of the interpreting cardiac imager for the evaluation of luminal stenosis associated with calcified lesions or stents. This resulted in a change of CAD-RADS score in 6 patients, with three patients being downgraded from 4 (70–99 % stenosis) to 3 (50–69 % stenosis). This is a significant difference since a CAD RADS score of 4 warrants cardiac catheterization or functional imaging, whereas a category of 3 may not need it. A combination of 50 and 100 keV VMIs may be useful for the interpretation of CCTAs, providing good image quality, contrast enhancement and lower blooming artifacts.
There are a few limitations that warrant mention. The scans in this study were obtained without the use of nitroglycerine due to absence of nursing support, which might have decreased the quality of some of our studies. However, the overall image quality was not substantively compromised, which is likely due to a combination of high temporal resolution of the scanner and selection of patients with heart rates < 70 bpm. Further improvement may be expected in routine clinical use with medications. We decreased the contrast dose by 50% in the low-dose cohort, but this can be personalized to the individual patient for future studies. We did not evaluate other ME images such as iodine maps and VNC. Calcium blooming can also be decreased with the use of high-resolution mode for cardiac imaging in the PCD-CT system (13,14,36). However, this mode was not available on the PCD-CT system at the time of this investigation. A combined high-resolution multi-energy mode is currently available for non-ECG-gated studies. We demonstrated changes in CAD-RADS categories based on calcium blooming reduction, but we do not have a reference technique for establishing the ground truth. Future studies will evaluate the utility of additional ME images and the benefits of high-resolution mode and correlation with a reference standard.
In conclusion, the high-pitch helical multi-energy mode of PCD-CT provided high-quality images with good contrast and low motion at low radiation and iodinated contrast doses. The quality of 50 keV VMI is superior to 120 kV images and can be used for routine clinical interpretation. The 50 keV VMIs have high iodine signal, which can improve low-contrast dosage studies. 100 keV VMIs improved the diagnostic confidence in the assessment of luminal stenosis associated with calcific plaques and stents.
Conflicts of Interest and Source of Funding:
Research reported in this publication was supported in part by the National Institute of Biomedical Imaging and Bioengineering of the National Institutes of Health under award number R01 EB028590. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The CT system used in this study was provided to Mayo Clinic through a grant to the author’s institution. Research support for Cynthia McCollough and Joel Fletcher is provided to Mayo Clinic from Siemens Healthcare GmbH. The other authors have no relevant conflicts of interest to disclose.
REFERENCES
- 1.Gulati M, Levy PD, Mukherjee D, et al. 2021. AHA/ACC/ASE/CHEST/SAEM/SCCT/ SCMR Guideline for the Evaluation and Diagnosis of Chest Pain: Executive Summary: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2021;144(22): e368–e454. [DOI] [PubMed] [Google Scholar]
- 2.Moss AJ, Williams MC, Newby DE, Nicol ED. The Updated NICE Guidelines: Cardiac CT as the First-Line Test for Coronary Artery Disease. Current Cardiovascular Imaging Reports. 2017;10(5):15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Davenport MS, Perazella MA, Yee J, et al. Use of intravenous iodinated contrast media in patients with kidney disease: Consensus statements from the American College of Radiology and the national Kidney Foundation. Radiology 2020; 294 (3): 660–8. [DOI] [PubMed] [Google Scholar]
- 4.Grist TM, Canon CL, Fishman EK, et al. Short-, mid-, and long-term strategies to manage shortage of iohexol 2022; 204:289–293. [DOI] [PubMed] [Google Scholar]
- 5.Raju R, Thompson AG, Lee K, et al. Reduced iodine load with CT coronary angiography using dual-energy imaging: a prospective randomized trial compared with standard coronary CT angiography. J Cardiovasc Comput Tomogr 2014; 8 (4): 282–288. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rotzinger DC, Si-Mohamed SA, Yerly J, et al. Reduced-iodine dose dual-energy coronary CT angiography: qualitative and quantitative comparison between virtual monochromatic and polychromatic images. Eur Radiol 2021; 31 (9): 7132–7142 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Zhang LJ, Qi L, De Cecco CN, et al. High-pitch coronary CT angiography at 70 kVp with low contrast media volume. Medicine 2014; 93 (22):e92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kok M, Mihl C, Hendriks BMF, et al. Optimizing contrast media application in coronary CT angiography at lower tube voltage: evaluation in a circulation phantom and sixty patients. Eur J Radiol. 2016; 85:1068–1074. [DOI] [PubMed] [Google Scholar]
- 9.van Hamersvelt RW, Eijsvoogel NG, Mihl C, et al. Contrast agent concentration optimization in CTA using low tube voltage and dual-energy CT in multiple vendors: a phantom study. Int J Cardiovasc Imaging. 2018; 34:1265–1275. [DOI] [PubMed] [Google Scholar]
- 10.Rajendran K, Petersilka M, Henning A, et al. First clinical photon-counting detector CT system: Technical evaluation. Radiology 2022; 303 (1):130–138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Leng S, Yu Z, Halaweish A, et al. Dose-efficient ultrahigh-resolution scan mode using a photon counting detector computed tomography system. J Med Imaging (Bellingham) 2016; 3 (4): 043504. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Leng S, Bruesewitz M, Tao S, et al. Photon-counting detector CT: System design and clinical applications of an emerging technology. Radiographics 2019; 39 (3): 729–743. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Mergen V, Sartoretti T, Baer-Beck M, et al. Ultra-high resolution coronary CT angiography with photon-counting detector CT: Feasibility and image characterization. Invest Radiol 2022; 57 (12): 780–788. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Si-Mohamed SA, Boccalini S, Lacombe H, et al. Coronary CT angiography with photon-counting CT: First-in-human results. Radiology 2022; 303 (2): 303–313. [DOI] [PubMed] [Google Scholar]
- 15.Boccalini S, Si-Mohamed SA, Lacombe H, et al. First-in-human results of computed tomography angiography of coronary stent assessment with a spectral photon counting computed tomography. Invest Radiol 2022; 57 (4): 212–221. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Mergen V, Eberhard M, Manka R, et al. First in-human quantitative plaque characterization with ultra-high-resolution coronary photon-counting CT angiography. Front Cardiovasc Med 2022; Sep 6:9:981012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Mergen V, Sartoretti T, Klotz E, et al. Extracellular volume quantification with cardiac late enhancement scanning using dual-source photon-counting detector CT. Invest Radiol 2022; 57 (6): 406–411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Mergen V, Ried E, Allmendinger T, et al. Epicardial adipose tissue attenuation and fat attenuation index: Phantom study and in vivo measurements with photon-counting detector CT. AJR Am J Roentgenol 222; 218 (5):822–829. [DOI] [PubMed] [Google Scholar]
- 19.Ahmed Z, Campeau D, Gong H, et al. Technical Note: High-pitch, high temporal resolution, multi-energy cardiac imaging on a dual-source photon-counting detector CT. Medical Physics 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Emrich T, O’Doherty J, Schoepf UJ, et al. Reduced iodinated contrast media administration in coronary CT angiography on a clinical photon-counting detector CT System: A phantom study using dynamic circulation model. Invest Radiol 2022; 10.1097/RLI.0000000000000911 [DOI] [PubMed] [Google Scholar]
- 21.Euler A, Higashigaito K, Mergen V, et al. High-pitch photon-counting detector computed tomography angiography of the aorta: Intraindividual comparison to energy-integrating detector computed tomography at equal radiation dose. Invest Radiol 2022; 57 (2): 115–121. [DOI] [PubMed] [Google Scholar]
- 22.Symons R, Sandfort V, Mallek M, Ulzheimer S, Pourmorteza A. Coronary artery calcium scoring with photon-counting CT: first in vivo human experience. Int J Car [DOI] [PubMed] [Google Scholar]
- 23.Secchi F, De Cecco CN, Spearman JV, et al. Monoenergetic extrapolation of cardiac dual energy CT for artifact reduction. Acta Radiol. 2015;56(4):413–8. [DOI] [PubMed] [Google Scholar]
- 24.Van Hedent S, Große Hokamp N, Kessner R, et al. Effect of Virtual Monoenergetic Images from Spectral Detector Computed Tomography on Coronary Calcium Blooming. J Comput Assist Tomogr. 2018;42(6):912–8. [DOI] [PubMed] [Google Scholar]
- 25.Huang X, Gao S, Ma Y, et al. The optimal monoenergetic spectral image level of coronary computed tomography angiography on a dual-layer spectral detector CT with half-dose contrast media. Quantitative Imaging in Medicine and Surgery 2020; 10 (3): 592–603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Cury RC, Abbara S, Achenbach S, et al. CAD-RADS(TM) Coronary artery disease - Reporting and Data System. An expert consensus document of the Society of Cardiovascular Computed Tomography (SCCT), the American College of Radiology (ACR) and the North American Society for Cardiovascular Imaging (NASCI). Endorsed by the American College of Cardiology. J Cardiovasc Comput Tomogr. 2016;10(4):269–81. [DOI] [PubMed] [Google Scholar]
- 27.Cury RC, Leipsic J, Abbara S, et al. CADRADS 2.0– 2022 coronary artery disease- Reporting and Data System an expert consensus document of the Society of Cardiovascular Computed Tomography (SCCT), the American College of Cardiology (ACC), the American College of Radiology (ACR) and the North American Society of Cardiovascular Imaging (NASCI). J Cardiovasc Comput Tomogr., July 7, 2022. [DOI] [PubMed] [Google Scholar]
- 28.Achenbach S, Marwan M, Ropers D, et al. Coronary computed tomography angiography with a consistent dose below 1 mSv using prospectively electrocardiogram-triggered high-pitch spiral acquisition. Eur Heart J. 2010;31(3):340–6. [DOI] [PubMed] [Google Scholar]
- 29.Kröpil P, Rojas CA, Ghoshhajra B, et al. Prospectively ECG-triggered high-pitch spiral acquisition for cardiac CT angiography in routine clinical practice: initial results. J Thorac Imaging. 2012;27(3):194–201. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Smettei OA, Sayed S, Abdullah MAH, et al. Ultra-fast, low dose high pitch (FLASH) versus prospectively gated coronary computed tomography angiography: Comparison of image quality and patient radiation exposure. J Saudi Heart Assoc. 2018;30(3):165–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Liu Q, Wang Y, Qi H, et al. Exploring the best monochromatic energy level in dual energy spectral imaging for coronary stents after percutaneous coronary intervention. Scientific Reports 2021; 11:17576. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Wang T, Ghadivel BB, Beitler JJ, et al. Optimal virtual monoenergetic image in “Twinbeam” dual-energy Ct for organs-at-risk delineation based on contrast-noise-ratio in head-and-neck radiotherapy. J Appl Clin Med Phys 2019; 20 (2):121–128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Zeng Y, Geng D, Zhang J. Noise-optimized virtual monoenergetic imaging technology of the third-generation dual-source computed tomography and its clinical applications. Quant Imaging Med Surg 2021; 11 (11): 4627–4643. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Kalisz K, Halliburton S, Abbara S, et al. Update on Cardiovascular Applications of Multienergy CT. Radiographics. 2017;37(7):1955–74. [DOI] [PubMed] [Google Scholar]
- 35.Rajiah P, Parakh A, Kay F, et al. Update on Multienergy CT: Physics, Principles, and Applications. Radiographics. 2020;40(5):1284–308. [DOI] [PubMed] [Google Scholar]
- 36.Latina J, Shabani M, Kapoor K, et al. Ultra-High-Resolution Coronary CT Angiography for Assessment of Patients with Severe Coronary Artery Calcification: Initial Experience. Radiol Cardiothorac Imaging 2021;3(4): e210053. [DOI] [PMC free article] [PubMed] [Google Scholar]