RPS 2103 - Cardiac imaging biomarkers for prognosis and clinical outcomes

Author Block: B. Bogner, M. Jung, M. Reisert, J. Maushagen, S. Rospleszcz, C. L. Schlett, F. Bamberg, J. Weiß; Freiburg Im Breisgau/DE
Purpose: To determine whether MRI-derived visceral-to-subcutaneous adipose tissue (VAT/SAT) ratio predicts incident diabetes and major adverse cardiovascular events (MACE) beyond other risk factors in metabolically healthy non-obese (MHN) individuals.
Methods or Background: We analyzed UK Biobank participants who were MHN (absence of hyperlipidemia or hypertension, diabetes, and BMI <30 kg/m²). 3D VAT/SAT volumes (dm3) were extracted from Dixon whole-body MRI using a validated deep learning framework. Sex-specific VAT/SAT ratio cutoffs (≥0.15 for females, ≥0.35 for males) were derived from C-index optimization for incident cardiometabolic outcomes. Cox proportional hazards models assessed associations with 1) incident diabetes and 2) MACE, with stepwise adjustment for age, sex, smoking, waist circumference (WC), and BMI. Multivariable nested models with and without VAT/SAT ratio were compared to test for added value beyond other factors. Net reclassification improvement (NRI) compared risk classification by VAT/SAT versus WC.
Results or Findings: Among 22,040 participants (mean age 64.6±7.8 years, 53% female), over a median 4.2-year follow-up, 251 diabetes (1.14%) and 297 MACE (1.35%) events occurred. High VAT/SAT ratio was independently associated with diabetes (aHR 1.77, 95% CI 1.34-2.33, p<0.001) and MACE (aHR 1.30, 95% CI 1.02-1.66, p=0.037) after adjustment for age, sex, smoking, BMI, and WC. Adding VAT/SAT to fully adjusted models significantly improved discrimination for diabetes (C-index 0.723 vs. 0.715, p<0.001) and MACE (C-index 0.694 vs. 0.690, p=0.036). VAT/SAT outperformed WC with significant NRI for diabetes (NRI 0.102, p=0.010) and MACE (NRI 0.088, p=0.013).
Conclusion: VAT/SAT ratio independently predicts diabetes and MACE in MHN individuals and demonstrates superior risk classification over WC, highlighting the potential of body composition analysis for personalized risk assessment in apparently healthy individuals.
Limitations: The generalizability is limited by the UK Biobank's selection bias, the VAT/SAT cut-off derivation, and the focus on MHN individuals.
Funding for this study: None.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: University of Freiburg

Author Block: A. M. Bucher¹, M. Neitzel¹, C. Ehrengut², T. S. Emrich³, R. Klöckner⁴, M. M. Sieren⁴, M. A. Fink⁵, F. G. Meinel⁶, A. S. Surov⁷; ¹Frankfurt/DE, ²Leipzig/DE, ³Mainz/DE, ⁴Lübeck/DE, ⁵Heidelberg/DE, ⁶Rostock/DE, ⁷Minden/DE
Purpose: To investigate the prognostic value of CT-defined coronary calcifications in COVID-19 patients using the semiquantitative Weston score.
Methods or Background: This retrospective multicenter study included 541 PCR-confirmed COVID-19 patients (176 female, mean age 61.2±15.6 years) from 11 German university hospitals. Coronary calcifications were assessed on chest CT using the Weston score (0-12 points), evaluating four coronary segments on a 4-point scale. Primary endpoint was 30-day mortality. Statistical analysis included univariable and multivariable regression.
Results or Findings: Overall 30-day mortality was 21.2% (115/541). Mean Weston score was 3.0±3.6, with 23.7% showing no calcifications. Patients with fatal outcomes had significantly higher Weston scores (4.1±4.01 vs. 2.7±3.4, p=0.0007). While presence of calcifications showed association with mortality in univariable analysis (OR 1.68, 95%CI 1.08-2.59, p=0.01), this lost significance in multivariable analysis (p=0.49). However, Weston score as continuous variable remained significant in both univariable (OR 1.10, 95%CI 1.04-1.14, p<0.001) and multivariable analysis (OR 1.06, 95%CI 1.005-1.138, p=0.036) after adjusting for age and COVID-19 CT score.
Conclusion: The extent of coronary calcifications, quantified by Weston score, is an independent prognostic factor for 30-day mortality in COVID-19 patients. The degree of calcification provides greater prognostic value than mere presence. These findings support integrating coronary calcification assessment into COVID-19 risk stratification using routine chest CT without dedicated cardiac imaging.
Limitations: The limitation of the study is the retrospective study design.
Funding for this study: Funding was provided by the German Federal Ministry of Education and Research (BMBF), Project RACOON (01KX2021).
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Multicentric ethics approval was obtained by University Medicine Frankfurt (Reference 20-719jK).

Author Block: S. S. Dereli Bulut, G. Karamustafao; Istanbul/TR
Purpose: To evaluate whether virtual intravascular ultrasound (IVUS)-derived intimal thickness, calcium scoring, and positive remodeling on coronary CT can predict progression from CAD-RADS 0 to CAD-RADS 1, and to assess their relationship with cardiovascular risk factors.
Methods or Background: In this retrospective pilot study, 91 patients with CAD-RADS 0–1 (CAD-RADS 2.0)who underwent 128-slice CCTA (2021–2025) were analyzed. Intimal thickness was measured on reformatted axial planes using virtual IVUS,defined as (outer diameter–inner diameter)/2. Two radiologists performed independent measurements, with interobserver agreement assessed by ICC. Calcium scoring (Smart Score) and positive remodeling (Plaque ID) were also evaluated, and risk factors (hypertension, diabetes, smoking, LDL) retrieved from medical records.
Results or Findings: CAD-RADS 0 (n=45: 44.8 ± 11.6 years, 46.7% male, BMI 26.9 ± 3.7) and CAD-RADS 1 (n=46: 51.6 ± 9.9 years, 52.6% male, BMI 28.4 ± 2.8) differed significantly. Intimal thickness was greater in CAD-RADS 1, especially mid LAD, LCX, and RCA (p<0.001). Calcium scores were higher, particularly LAD (0.0 vs 21.7; p<0.001). Positive remodeling occurred in 93% vs 42% (p<0.001). Intimal thickness correlated with calcium scores, strongest in mid LAD (r=0.71, p<0.001). Mid LAD thickness was the strongest predictor (p=0.01); the combined model (LAD-mid + LDL + HT) yielded AUC=0.96 (98% sensitivity, 89% specificity).
Conclusion: IVUS-derived intimal thickening, particularly in the mid LAD, may serve as an early imaging marker of CAD-RADS progression. Combined with calcium scoring and remodeling, IVUS provides value for early risk stratification. These preliminary results are hypothesis-generating and need confirmation in larger prospective cohorts.
Limitations: This single-center pilot study had a modest sample size, included only CAD-RADS 0–1 patients without longitudinal follow-up, and may still be subject to measurement variability despite good reproducibility. Other biomarkers such as inflammation or genetics were not assessed.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This retrospective study was approved by the Ethics Committee of Umraniye Training and Research Hospital (Approval number: E-54132726-000-282160058)

Author Block: F. Piccione¹, D. Vignale², A. Colombo², A. Palmisano², C. Gnasso², G. Amaro², A. Esposito²; ¹Turin/IT, ²Milan/IT
Purpose: Assessing the prognostic value of myocardial strain and extracellular volume fraction (ECV) measured by pre-procedural CT in patients with severe aortic stenosis (AS) undergoing transcatheter aortic valve implantation (TAVI).
Methods or Background: From October 2020 to December 2023 patients were screened using as exclusion criteria aortic regurgitation, peri-procedural complications, incomplete strain/ECV data.
Global strain (longitudinal, radial, and circumferential) was quantified using dedicated software (CVI42 v6.3.0). ECV was measured manually and automatically (in-house built software) in the interventricular septum as the ratio of the differential attenuation of myocardium and blood pool before and after contrast administration. Primary endpoint: death or HF hospitalization after a follow-up of 12+ months.
Outcomes were analyzed with Kaplan–Meier survival and Cox regression.
Results or Findings: In 241 patients (median age 82 (IQR 78–86) years, 53.9% females) primary endpoint occurred in 18.3%. Patients with events had higher ECV (31.6% vs 27.6% manual, p<0.001; 31.7% vs 28.7% automatic, p=0.003), more impaired longitudinal (-13.0% vs -15.3%, p=0.035) and radial strain (18.1% vs 23.2%, p=0.030). Survival curves identified ECV cut-offs of 30.6% (manual) and 28.9% (automatic), both associated with worse event-free survival (p<0.001). In multivariate analysis, ECV remained an independent predictor (HR 2.12 manual; HR 2.45 automatic), while strain lost significance (HR 1). Patients with elevated ECV and impaired longitudinal strain had the highest risk (HR up to 3.04). Classical low-flow low-gradient AS showed the worst remodeling (ECV 33–34%, GLS –9%).
Conclusion: Pre-procedural CT-derived ECV represents an independent and robust predictor of adverse prognosis after TAVI.
Combined ECV–strain assessment may provide a more comprehensive characterization of myocardial remodeling and improve risk stratification. This approach offers the additional advantages of being automatable, reproducible, and objective. Larger prospective studies are warranted to confirm its clinical applicability.
Limitations: Small single-center cohort.
Funding for this study: No funding
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Retrospective study, based on prospectively collected data, and was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants.

Author Block: P. Kamieniarz, A-S. Plietz, L. V. Urbach, M. Taupitz, L-A. Schaafs; Berlin/DE
Purpose: Chronic kidney disease (CKD) is a global health burden with increasing prevalence. Those effected have an increased risk for cardiovascular disease (CVD) and higher mortality. Myocardial strain derived from cardiac MRI (cMRI) is shown to be an important prognostic marker in a variety of CVD. We aimed to assess left ventricular (LV) strain in patients with CKD and compare them to a healthy control group.
Methods or Background: In this single centre, retrospective study we included 288 patients with CKD in all stages of disease (G1-G5) and 74 healthy controls. All patients underwent comprehensive cMRI. Functional analysis was performed using CVI 42 (Circle Vascular Imaging, Calgary, Canada). For strain analysis Segment® (Medviso, Lund, Sweden) was used. Differences between groups were assessed using ANOVA.
Results or Findings: LV ejection fraction (59.4±6.4 vs. 47.5±16.1; 51.0±16.0; 46.1±17.4; 48.8±20.8; 49.0±18.2; p<0.001 ), LV global longitudinal strain (-18.5±3.1 vs. -13.9±5.5; 15.3±4.9; -13.7±5.8; -11.8±5.8; -11.5±4.6; -11.3±4.6; p<0.001), LV global circumferential strain (-21.8±3.5; -16.4±7.7; -19.0±6.1; -17.2±6.8; -15.9±7.2; -15.9±6.4; p<0.001) and LV global radial strain (50.3±9.8 vs. 37.4±19.5; 45.7±18.6; 39.8±18.5; 34.8±19.4; 40.0±18.7; 36.2±18.1; p<0.001) were statistically significant impaired in patients with CKD stages G1 – G5 compared to healthy controls.
Conclusion: Patients with CKD had impaired LV strain with the biggest differences observed in Groups G1-G3a. In groups G3b – G5 a gradual recovery of LV strain parameters was noticed. With LV strain severely impaired in patients with CKD, LV ejection fraction was only mildly reduced. Influencing factors such as age, sex and volume status need to be analysed in future studies.
Limitations: The limitations of this study are the retrospective design, variance in comorbidities and that groups could not be matched for sex and age.
Funding for this study: This research was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) - Grant No. 372486779 (SFB 1340/2)
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Ethik-Kommission der Charité (EA2/237/21)

Author Block: R. Aroasio¹, L. Tampellini¹, F. Minonne¹, S. Tibaldi¹, G. Balladori¹, L. Blasi¹, M. Moretti¹, M. Gatti¹, R. Faletti²; ¹Turin/IT, ²Torino/IT
Purpose: Evaluate the impact of coronary CT angiography (CCTA) on cardiovascular risk stratification and the prevention of MACE in liver transplant (OLT) candidates.
Methods or Background: Single-center retrospective study on 140 consecutive OLT candidates enrolled between April 2021 and October 2024, all with ≥ 2 coronary risk factors and undergoing CCTA. Positive exams, when clinically indicated, were further investigated with invasive coronary angiography (ICA) and possible revascularization; negative CCTAs were considered true negatives (TN) if supported by MACE-free clinical follow-up ≥ 6 months. Primary endpoint: MACE within 30 days after OLT.
Results or Findings: A total of 140 OLT candidates (78% male) were consecutively enrolled; 55% were > 65 years old and 32% had NASH cirrhosis. CCTA excluded significant stenosis in 95/140 patients (68%) and detected it in 27/140 (19%); 23/27 (85%) underwent ICA, with confirmation and PCI in 14/23 (61%). Using the TN6m definition (100 patients), the 2 × 2 matrix showed 14 true positives, 9 false positives, 0 false negatives, and 100 TN, with sensitivity and NPV 100%, specificity 91%, and PPV 60.9%. Among 91 transplanted patients, no MACE occurred within 30 days (incidence 0%). These patients accumulated 27,097 days (74.2 patient-years) of event-free follow-up; estimated incidence would be 1.1% assuming a single MACE and 5.5% with five events.
Conclusion: With 100% sensitivity and NPV at 6 months, CCTA proves to be a highly reliable screening tool to exclude significant coronary artery disease in OLT candidates with cardiovascular risk factors. The absence of MACE within 30 days post-transplant highlights the effectiveness of the CCTA-first strategy in perioperative protection. Prospective multicenter studies are needed for long-term validation.
Limitations: Retrospective and single-center design, relatively small sample size, short- to mid-term clinical follow-up.
Funding for this study: No funding for this study
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: No additional information

Author Block: Y. Hu¹, W. Deng², J. Zhang²; ¹Yan'cheng, Jiangsu/CN, ²Shanghai/CN
Purpose: Epicardial adipose tissue (EAT) is an important biomarker in atrial fibrillation (AF). This study evaluated whether EAT histogram parameters from virtual noncontrast and contrast-enhanced dual-layer spectral CT predict AF recurrence after ablation.
Methods or Background: In this prospective study, patients undergoing first-time AF ablation (Jan 2021–Jul 2024) underwent preoperative dual-energy cardiac CT. EAT on contrast-enhanced images (-190 to -30 HU) was segmented using a deep learning model. VNC images were reconstructed, and density difference maps were generated. Regional histogram parameters were extracted. Continuous variables were compared using Mann–Whitney U test, and Cox models evaluated predictors of AF recurrence.
Results or Findings: A total of 492 patients (mean age, 66 years; 316 men) were enrolled (paroxysmal AF, 54%; persistent AF, 46%). The median follow-up was 22 months. EAT volume was significantly greater in patients with persistent AF compared with those with paroxysmal AF (144 cm³ [IQR, 105–184] vs 126 cm³ [IQR, 100–154]; P < .001). Multiple EAT histogram parameters differed significantly between groups, including Max, Mean, Median, Std, Q1, Q3, Variance, IQR, 90th percentile, MAD, Skewness, Energy, Total Energy, RMS, Entropy, and Uniformity. During follow-up through July 2025, AF recurrence occurred in 73 patients (15%). Among EAT regional histogram parameters, entropy was significantly associated with AF recurrence (hazard ratio [HR], 2.0; 95% CI: 1.2–3.3; P < .01). In multivariable Cox regression, EAT entropy remained an independent predictor after adjustment for age, sex, AF phenotype, and EAT volume (HR, 1.97; 95% CI: 1.2–3.3; P = .013).
Conclusion: DLCT-based EAT histogram analysis, especially regional entropy, offers novel biomarkers for AF recurrence, supporting enhanced pre-ablation risk stratification.
Limitations: One-center study
Funding for this study: This study is supported by The National Key Research and Development Program of China (Grant No.: 2021YFF0501402), National Natural Science Foundation of China (Grant No.: 82471982, 82271990), Shanghai Municipal Science and Technology Commission Discipline Leader Project (Grant No.: BJKJ2024052), Shanghai Health Commission Discipline Leader Project (Grant No.: 2022XD031), and Medical research project of Yancheng Municipal Health Commission (Grant No.: YK2024225).
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The institutional review board of each participating hospital approved this prospective study (2018-014)

Author Block: E. Zuhri¹, E. Elen¹, S. J. S. Gardezi², D. Kumar², G. Marcel¹, C. A. Atmadikoesoemah¹; ¹Jakarta/ID, ²Shanghai/CN
Purpose: Atrial fibrillation (AF) is one of the most common cardiac arrhythmias and poses a major challenge in the management of cardiovascular disease. Catheter ablation (CA) is an established therapeutic option for patients with drug-resistant AF. However, recurrent AF following CA remains a major clinical challenge. Recent studies suggest that peri-coronary adipose tissue (PCAT), particularly surrounding the left circumflex artery (LCx), may contribute to AF recurrence after CA. This study aims to identify this potential factor for prediction of recurrent AF after CA.
Methods or Background: Patients with recurrent AF after CA were retrospectively enrolled. We investigated the predictive value of PCAT and other clinical factors for AF recurrence after CA. PCAT was measured using AI-based tool (uAI Fat Attenuation Index) developed by United Imaging Intelligence. Bivariate and multivariate logistics regression analyses were performed
Results or Findings: During 1‐year follow‐up, 18 of 33 patients experienced AF recurrence. The multivariable analysis revealed that PCAT of the LCx was independently associated with AF recurrence. PCAT of the LCx showed a significant predictive value, with a cut-off value of >−86.5 HU, yielding 78% sensitivity and 67% specificity. The area under the receiver operating characteristic curve (AUC) was 0.720, indicating good discriminative ability. Other factors-including age, sex, obesity, hypertension, diabetes, dyslipidemia, smoking, chronic kidney disease, heart failure, reduced TAPSE, valvular heart disease, coronary artery disease, and use of antiarrhythmic drugs (AADs, Class II/III)-were not significantly associated AF recurrence.
Conclusion: PCAT of the LCx was associated with AF recurrence after CA and may have important clinical value in predicting recurrence.
Limitations: Larger sample are needed to validate these findings and further refine risk prediction models for AF recurrence.
Funding for this study: None
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: L. Fulceri, L. Taddei, A. Marcucci, L. Faggioni, D. Cioni, E. Neri, G. Aquaro; Pisa/IT
Purpose: To evaluate the prognostic role of FR compared to late gadolinium enhancement (LGE) in subjects with NDLVC, preserved LVEF and frequent premature ventricular complexes (PVC).
Methods or Background: Cardiac magnetic resonance was performed in 900 consecutive patients with frequent PVC, preserved LVEF. Patients with diagnostic criteria for other cardiomyopathies, channelopathies and ischemic disease were excluded. A clinical follow-up was performed and major ventricular arrhythmias (MVA), including sudden cardiac death, appropriated ICD intervention, and sustained VT, were considered endpoints.
Results or Findings: The final population included 876 patients. Overall non-ischemic LGE was found in 101 (12%) patients and FR in 38 (4%): 87 (10%) presenting LGE without FR, 14 (1,6%) FR without LGE and 24 (3%) LGE with FR. After a median follow-up of 4 (3-6) years, MVA occurred in 43 patients. At Kaplan-Meier survival curve analysis (figure 1) the presence of LGE and/or FR was associated with worse prognosis than their absence. The group of patients with worse prognosis was that with the combined LGE+FR. A similar prognosis was found in patients with LGE alone and in those with FR alone (p=0.19). Patients with LGE+FR had worse prognosis than those with LGE alone (P=0.01). At univariate analysis non-sustained VT (NSVT), syncope, LGE, FR and LGE+FR were associated to MVA. At multivariable logistic regression analysis the combination LGE+FR was an independent predictor of MVA together with history of NSVT and syncope.
Conclusion: In NDLVC with preserved LVEF, the combination of FR+LGE was a stronger prognostic marker than the presence of LGE alone. FR and LGE had similar prognostic role when found separately.
Limitations: Relatively small population
Funding for this study: Istitutional funds of the University of Pisa
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: W. Zhang, X. Xu, W. Peng, K. Shi, X. Lyu, C. Xia, Z. Li; Chengdu/CN
Purpose: To evaluate the prognostic value of pulmonary transit time (PTT) assessed by cardiac magnetic resonance (CMR) in patients with dilated cardiomyopathy (DCM).
Methods or Background: Recent studies confirm CMR first-pass perfusion imaging offers a convenient, non-invasive method to measure PTT, a marker of global cardiopulmonary circulation. However, its prognostic value in DCM patients has not been established. This retrospective study enrolled patients with DCM who underwent CMR at our institution between July 2019 and January 2022. PTT was measured as the peak-to-peak time between the contrast time-intensity curves of the right and left ventricles in first-pass perfusion imaging. Cox regression analysis was performed to assess the association between PTT and a composite endpoint comprising all-cause mortality, heart failure rehospitalization, and heart transplantation.
Results or Findings: A total of 140 patients (71.4% males, mean age 52±13 years) were included and followed for a median of 4.6 years (interquartile range, 4.1–5.4 years). During this period, 70 patients (50.0%) reached the endpoint. Multivariable linear regression identified PTT as an independent determinant of left ventricular (LV) remodeling and dysfunction, as it correlated with increased LV end-diastolic dimension (β=0.431), reduced LV ejection fraction (β=-0.391), and impaired LV global longitudinal peak strain (β=0.276). Receiver operating characteristic curve analysis, using the maximization of Youden's index, identified an optimal PTT cut-off of 10.7 seconds for endpoint prediction. Kaplan–Meier survival analysis revealed that patients with a PTT greater than 10.7 seconds had a significantly higher risk of experiencing the composite endpoint (log-rank p<0.001). In multivariable Cox regression analysis, PTT was an independent predictor of the endpoint (adjusted hazard ratio: 1.072 per second; 95% confidence interval: 1.006–1.144; p=0.033).
Conclusion: PTT derived from CMR independently predicted adverse cardiovascular outcomes in patients with DCM.
Limitations: Not applicable
Funding for this study: Not applicable
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: