RPS 2403 - Photon-counting CT and contrast media optimisation

Author Block: P. Levoni¹, F. Catapano¹, C. Lisi¹, A. Laghi¹, M. Francone², L. S. Politi¹; ¹via Rita Levi Montalcini 4, 20072 Pieve Emanuele, Milan/IT, ²Sant'Andrea Hospital, via di Grottarossa 1035-1039, 00189 Rome/IT
Purpose: Photon-counting detector CT (PCD-CT) has been proposed to reduce radiation exposure compared with energy-integrating detector CT (EID-CT). Nonetheless, data on patients’ radiation dose in cardiac applications are still limited. The aim of the present study is to quantify and analyse radiation dose across three cardiac PCD-CT protocols, and to compare the results with a standard cardiac EID-CT protocol.
Methods or Background: A cohort of 268 patients who underwent coronary CT angiography between December 2024 and July 2025 was retrospectively included. Each patient underwent one of four acquisition protocols: three PCD-CT protocols - Spectral; Ultra-High Resolution (UHR); UHR Spectral - and one EID-CT protocol. For each acquisition volume, CT dose index (CTDIvol), dose length product (DLP), and size-specific dose estimate (SSDE), were retrieved, and total DLP was calculated. Statistical analysis included descriptive assessment and Kruskal–Wallis H-tests to compare dose distributions across protocols.
Results or Findings: Mean±SD total DLP values were 653±192 mGy·cm for the Spectral protocol, 947±233 mGy·cm for the UHR protocol, 1373±316 mGy·cm for the UHR Spectral protocol, and 390±114 mGy·cm for the EID-CT protocol. Kruskal-Wallis H-tests showed that each pair of distributions is different with statistical relevance. The median value of total DLP for the UHR Spectral protocol (1356 mGy·cm) results greater than the corresponding Italian Dose Reference Level (1200 mGy·cm).
Conclusion: Cardiac PCD-CT protocols resulted in higher radiation doses compared with a conventional cardiac EID-CT protocol. Among PCD-CT protocols, UHR Spectral acquisitions frequently exceeded national reference limits. These findings suggest that PCD-CT protocols with default high-dose settings should be reserved for cases where maximal image quality is strictly required.
Limitations: The patient cohort undergoing the UHR protocol included a limited number of 20 patients.
Funding for this study: Funding was provided by the National Plan for NRRP Complementary Investments (PNC, established with the decree-law 6 May 2021, n. 59, converted by law n. 101 of 2021) in the call for the funding of research initiatives for technologies and innovative trajectories in the health and care sectors (Directorial Decree n. 931 of 06-06-2022) - project n. PNC0000003 - AdvaNced Technologies for Human-centrEd Medicine (project acronym: ANTHEM). This work reflects only the authors’ views and opinions, neither the Ministry for University and Research nor the European Commission can be considered responsible for them.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: The study is retrospective.

Author Block: E. Muscogiuri, T. Busselot, E. Di Dedda, S. Dymarkowski, H. Bosmans; Leuven/BE
Purpose: Compare coronary artery calcium score (CACS) from Photon Counting Computed Tomography (PCCT) images, acquired with contrast-enhanced ECG-gated and non-contrast, non-gated scan protocols, and reconstructed with different algorithms: “Pure Calcium” (PC) and “Virtual non-contrast” (VNC).
Methods or Background: Patients undergoing a PCCT scan before Transcatheter Aortic Valve Implantation were included. For each patient, CACS was measured in five datasets with dedicated software: “true” non-contrast ECG-scan (TNC, reference); PC and VNC reconstructions of the ECG-gated (PCGATED, VNCGATED) and non-gated contrast-enhanced scans (PCNON-GATED, VNCNON-GATED). Statistical pairwise comparisons per patient and per vessel (LM+LAD, LCX, RCA) were made using a Friedman test and Dunn’s multiple comparison test. Additionally, CACS was classified in risk categories (0; 0-10; 10-100; 100-400; >400) and compared using weighted Cohen's κ.
Results or Findings: Sixty-one patients were included (25 males, mean age 78,4 years). The analysis per patient showed that TNC CACS was not significantly different (p<0.05) from PCGATED, while it was significantly different from PCNON-GATED and VNC reconstructions. The analysis per vessel showed that CACS of LM+LAD calculated on PC reconstructions were not significantly different from TNC, while all the CACS of LCX were different from TNC. CACS of RCA calculated on PCGATED images were not significantly different from TNC. Classification comparison per CACS thresholds showed excellent agreement for PCGATED and PC-NONGATED (0.86 and 0.83, respectively) and a moderate to good agreement for VNCGATED and VNC-NONGATED (0.64 and 0.59, respectively).
Conclusion: The CACS measurements performed on PC reconstructions (particularly with ECG-gating) proved more reliable than those obtained from VNC reconstructions. However, it has to be noted that significant differences persist in the evaluation of CACS per vessel and in the risk stratification of patients among the different reconstructions.
Limitations: Small study population with the same disease.
Funding for this study: None
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Study number is S68949, approved on 20 June 2024.

Author Block: S. Tamietti¹, F. Piacentino², F. Fontana³, A. Gnesutta², M. Venturini⁴, C. Beltramini⁵, E. Macchi²; ¹Biella, BI/IT, ²Varese/IT, ³Vergiate/IT, ⁴Segrate/IT, ⁵Golasecca/IT
Purpose: To evaluated the image quality and plaque characterization in coronary CT angiography (CCTA) using dual-layer spectral CT (DL-SCT) with a reduced iodinated contrast bolus (35 mL) compared to the standard dose (70 mL). We focused on the role of virtual monoenergetic imaging (VMI) and Z-effective reconstructions.
Methods or Background: A cohort of 40 patients, with low-to-intermediate cardiovascular risk who underwent CCTA with DL-SCT (IQon Spectral CT Philips, Amsterdam, The Netherlands) were retrospectively divided in two groups based on contrast dose (70mL vs. 35mL of iomeron 400mg/mL). Quantitative parameters (contrast attenuation, signal-to-noise ratio [SNR], contrast-to-noise ratio [CNR]) were measured. Qualitative image analysis was performed by two radiologists. Plaque characterization was compared across conventional, VMI, and Z-effective reconstructions.
Results or Findings: In the reduced-contrast group, VMI at 50 keV achieved attenuation values comparable to standard acquisitions, while SNR and CNR were significantly higher (p < 0.05). Qualitative image scores were similar between groups; however, stenosis evaluation improved in the reduced-dose group due to fewer blooming artifacts in spectral reconstructions. Plaque analysis showed significant reduction in calcified plaque area and stenosis with VMI at 200 keV and Z-effective images, while VMI at 50 keV improved delineation of fibro-lipidic plaques and stenosis assessment (p < 0.05).
Conclusion: DL-SCT allows a 50% reduction in iodine load without compromising diagnostic quality. Low-energy VMI reconstructions maintain contrast enhancement, while VMI and Z-effective images enhance plaque characterization, supporting more accurate clinical management.
Limitations: The limitations of the study are the requirement for additional research on large, multicenter cohorts employing intra-patient protocols, in order to validate and improve future clinical practice.
Funding for this study: Not applicable.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: X. Cao, J. Li, X. Zhong, X. Liu; Guangzhou/CN
Purpose: To investigate the feasibility of reducing the contrast dose and the optimal keV of virtual monoenergetic images (VMIs) on dual layer spectral detector CT (DLCT) in infants and children with complex congenital heart disease (CHD).
Methods or Background: Prospectively enrolled 80 infants and children with complex CHD and underwent preoperative DLCT. Patients were randomly divided into two groups: group A (N=36, standard dose, 1.5~2 ml/kg) and group B (N=44, low dose, 1ml/kg). Conventional image (CI) and VMIs in range of 40-85 keV were reconstructed in group B while only CI was reconstructed in group A. Image quality was evaluated using qualitative scores and objective metrics including CT value, standard deviation (SD), signal-to noise ratio (SNR), and contrast-to-noise ratio (CNR) for the pulmonary artery and ascending aorta.
Results or Findings: Group B received 24% less contrast than group A [8.00 (5.00;11.00) ml vs 10.50 (8.00;21.00) ml, P<0.05]. VMI 40 keV in group B yielded the highest CT value, SNR, CNR of pulmonary artery and ascending aorta among different VIMs, and significantly outperformed CI in group A (P<0.05). Subjective evaluation score on VMI 40-50 keV of group B was significantly higher than CI of group A (P<0.05). Diagnostic accuracy for CHD at 40-50 keV in group B was 97.73%, significantly higher than group A (93.40%, P<0.05), while no significant difference in sensitivity and specificity.
Conclusion: VMI 40-50 keV in DLCT can significantly improve image quality, allowing contrast dose reduction in infants and children with complex CHD without compromising diagnostic performance.
Limitations: None
Funding for this study: None
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: None

Author Block: C. H. Park, S. Lee, J. W. Kim, J. E. Nam, E-Y. Choi, T. H. Kim; Seoul/KR
Purpose: Recent advances in low-kVp coronary CT angiography (CCTA) allow reduced iodine use, enabling the use of low-concentration iodine contrast media (ICM) while lowering radiation exposure. This study compared image quality and radiation dose between high- and low-concentration ICM in low-kVp, low-dose CCTA.
Methods or Background: 120 patients with suspected coronary artery disease were prospectively randomized to receive either high-concentration (Optiray, 350 mgI/mL) or low-concentration (Visipaque, 270 mgI/mL) contrast media. All CCTAs were performed using a 64-slice CT scanner with tube voltage and contrast volume adjusted to BMI. Quantitative (attenuation, noise, SNR, CNR) and qualitative (4-point scale) image quality were assessed. Acute and delayed adverse effects were monitored. Radiation dose was measured as dose-length product (DLP).
Results or Findings: No significant differences were found in demographic data, including heart rate. Attenuation (408.0 ± 67.9 HU vs. 424.5 ± 64.4 HU, p = 0.174), image noise, SNR, CNR, and qualitative image quality (mean scores: 3.34 vs. 3.46, p = 0.075) were comparable between groups. No acute adverse reactions occurred. Delayed adverse reactions were rare and non-severe, and resolved without sequelae. The radiation dose was not significantly different between the two groups (p = 0.842).
Conclusion: ICM dose reduction using low-concentration ICM is clinically feasible and provides comparable qualitative image quality in CCTA when using BMI-adapted low-kVp protocols in patients with suspected coronary artery disease.
Limitations: In this study, both groups used a low radiation dose protocol of either 100 kVp or 80 kVp. Subjects were randomly assigned to the two groups, resulting in no difference in body mass index. Tube voltage was determined by body mass index. Therefore, the radiation doses for both groups were similar, making it difficult to emphasize
Funding for this study: This study was funded by GE Healthcare. The funding sources had no role in the study design, data collection, data analysis, interpretation, or manuscript writing. The investigators maintained full responsibility for the content and the decision to submit the manuscript.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This prospective study was approved by the Ethics Committee and Institutional Review Board of Gangnam Severance Hospital.
All procedures followed established ethical guidelines. Informed consent was obtained from all participants.

Author Block: P. M. Tetteroo¹, C. Grinwis¹, L. R. Koetzier¹, D. Suchá¹, E. Langzam², N. R. van der Werf³, B. K. Velthuis¹; ¹Utrecht/NL, ²Haifa/IL, ³Best/NL
Purpose: To evaluate whether contrast-medium (CM)-dose in a pulmonary vein CT protocol can be safely reduced while maintaining diagnostic image quality using injection protocol adjustments and mono-energetic 55 keV reconstructions (MonoE55) on a dual-layer spectral CT (Philips CT7500).
Methods or Background: Sixty-eight patients were scanned with CM-dose concentrations progressively reduced from 100%, to 86%, 72%, 58% (N=17/group) with weight-adjusted (<80/≥80 kg) dual-phase contrast-injection protocols. Baseline contrast was 108/120 mL for <80/≥80 kg. Timing thresholds and injection speeds were modified from 200 HU in the descending aorta with 6/6.7 mL/s (<80/≥80 kg) for both 100%/86% CM-doses to 100 HU in the ascending aorta with 5/5.7 and 4.5/5.5 mL/s for 72%/58% CM-doses to optimize contrast timing. Subjective image quality (SIQ) was scored on MonoE55 and conventional images by one experienced, blinded cardiovascular radiologist using a 5-point Likert-scale for noise, contrast, attenuation, overall, and 3D volume. Objective image quality (OIQ) >200 HU was considered diagnostic and was assessed using 25 mm² regions-of-interest in four pulmonary veins, left atrial appendage, left atrium, and descending aorta. Data were compared to conventional 100% CM-dose images (reference) using Wilcoxon signed-rank or Mann–Whitney U tests for OIQ, and Wilcoxon signed-rank or chi-squared tests for SIQ.
Results or Findings: MonoE55 significantly improved SIQ across all CM-doses (p≤.026) compared with the reference, regardless of body weight, except for noise at 86% CM-dose, which was not significantly different (p=.295). MonoE55 significantly increased OIQ across all locations and for all CM-dose groups (p<.001). No MonoE55 attenuation measurements fell <200 HU.
Conclusion: CM-dose for pulmonary vein CT protocols can safely be reduced by 42% without compromising SIQ and OIQ when using spectral dual-layer MonoE55 reconstructions in combination with adjusted contrast timing and injection speed.
Limitations: Small sample size and single-reader design.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: R. Chen¹, H. Lu², A. Cernicanu³, J. Westenberg¹, M. Sevenster⁴, J. Keupp⁴, J. Meineke⁴, H. J. Lamb¹; ¹Leiden/NL, ²Amsterdam/NL, ³Eindhoven/NL, ⁴Hamburg/DE
Purpose: To develop and evaluate a free-breathing isotropic cardiovascular 3D cine MRI 1.5T framework that can assess cardiovascular anatomy and function without using contrast agent.
Methods or Background: A cartesian spiral(CASPR) bSSFP continuous k-space acquisition with 3D sagittal coverage of the whole heart and thoracic aorta was acquired twice, with 2.5mm and 2mm isotropic resolution. GPU-based reconstruction included retrospective binning of k-space profiles with respect to ECG and respiratory self-gating using a 1D feet-head projection. Conventional breath-hold 2D cine and 3D mDixon MR angiography sequences were acquired for comparison. Cardiovascular measurements were compared with Bland-Altman analysis and Intraclass Correlation Coefficient(ICC) for observer agreement(p<0.05).
Results or Findings: 24 volunteers were evaluated. Acquisition took 5-7minutes, with reconstruction about 5minutes per sequence. Compared to 2D cine, 3D cine demonstrated accurate measurement of left-ventricle ejection fraction, with a bias of 1.38% and limits of agreement(LoA) from −2.85% to 5.61% at 2.5mm, and a bias of 1.39% with LoA from −1.53% to 4.30% at 2.0mm. Aorta root sizes measured by 2D and 3D cine were comparable, showing a mean difference of 0.28mm2 with 95% LoA from -0.33mm2 to 0.89mm2 at 2.5mm, and a mean difference of 0.25mm2 with 95% LoA from -0.24mm2 to 0.74mm2 at 2.0mm. Other measurements showed strong agreement between 2D and 3D cine, with minimal mean differences and narrow limits of agreement. Intra- and interobserver for all the measurements showed at least moderate agreement, with ICC of 0.75 to 0.998.
Conclusion: Free-breathing isotropic cardiovascular 3D cine MRI using cartesian spiral sampling, without contrast agents, enables a simplified and time-efficient assessment of cardiovascular anatomy and function compared to 2D cine at 1.5T.
Limitations: There is a lack of studies including patients with arrhythmias and irregular breathing to validate image quality and quantification.
Funding for this study: No
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Medical Research Ethics Committee of Leiden University Medical Center

Author Block: V. Schmidt, L. Goertz, J. Tristram, R. A. Terzis, K. Kaya, D. Maintz, C. H. Gietzen, L. Pennig, J. P. Janssen; Köln/DE
Purpose: To compare REACT, a flow-independent 3D isotropic non-contrast MRA, with time-resolved contrast-enhanced MRA (4D CE-MRA) for postoperative assessment of the pulmonary arteries and right ventricular outflow tract, with emphasis on different implant types.
Methods or Background: In this retrospective single-center study, 53 patients with congenital heart disease underwent both 4D CE-MRA and REACT at 1.5 T. Three radiologists independently scored image quality (IQ), motion, and susceptibility on 5-point Likert scales for MPA, LPA, and RPA; diameters were measured and stenosis/dilatation recorded. Subgroup analyses were performed for stents, conduit/patch/valve (CPV), and no implant.
Results or Findings: REACT achieved higher overall IQ than 4D CE-MRA and significantly better motion scores in all vessels (all p < 0.001), while susceptibility scores were comparable (ns). The proportion of fully diagnostic studies (3/3 segments) was similar (REACT 77.4%, 41/53; 4D CE-MRA 83.0%, 44/53; McNemar ns). Diameter measurements showed excellent inter-reader agreement (ICC ≈ 0.89–0.95) and minimal bias between techniques; only the RPA was slightly smaller with REACT (mean difference –0.85 mm, p < 0.001), with Bland-Altman limits around ± 4 mm. In subgroup analyses, stented segments exhibited no IQ advantage of REACT and remained limited by susceptibility; motion scores remained superior with REACT, and diagnostic yield differences were not significant. In CPV and no-implant groups, REACT provided about +1 median IQ and motion score, with high diagnostic rates for both techniques.
Conclusion: REACT enables robust, contrast-free postoperative imaging of the pulmonary arteries with superior image quality and reduced motion artifacts compared to 4D CE-MRA, while maintaining highly reproducible diameter measurements. Stent segments remain a shared limitation; neither technique reliably evaluates stent lumen, and alternative modalities should be considered for targeted stent assessment.
Limitations: Non
Funding for this study: Non
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The local ethics committee approved this study.