BACKGROUND:Patients developed heart failure (HF) after acute myocardial infarction (AMI) have a high risk of mortality and rehospitalization, as an adverse consequence complicating AMI. Left ventricular ejection fraction (LVEF) at discharge does not accurately reflect cardiac functional status. Even patients with no HF on admission, and a normal LVEF at discharge are at considerable risk of developing HF. PURPOSE:This study aims to investigate the value of left atrial (LA) strain derived from cardiac magnetic resonance imaging (MRI) for further risk stratification for HF-related events in patients with ST-segment elevation myocardial infarction (STEMI) with various LVEF levels. MATERIALS AND METHODS:A total of 1531 patients with STEMI who underwent cardiac MRI were enrolled in this retrospective study. The endpoint is a composite of all-cause death, HF-related events, reinfarction, and unplanned revascularization. HF-related events were defined as unplanned hospitalization for HF and heart transplantation. RESULTS:Over a median follow-up of 50.9 months, 262 patients reached the endpoints, in which 104 patients developed HF-related events. In the adjusted analysis, diabetes, Killip class ≥ 2, microvascular obstruction, intramyocardial hemorrhage, ventricular aneurysm, infarct size, LA reservoir (hazard ratio [HR], 0.93 [95% CI, 0.91-.96]; P < 0.001), and conduit strain (HR, 0.94 [95% CI, 0.89-.99]; P = 0.02) were associated with HF-related events. The HF risk remained low in patients with relatively preserved LA reservoir function (≥23%), regardless of ejection fraction performance at discharge. Among populations with normal LVEF, and reduced LVEF, patients with LA reservoir strain <23% had a 2- to 5-fold increased risk of experiencing HF-related events. CONCLUSION:Reduced LA reservoir strain was independently associated with HF-related events in patients with STEMI, which can improve risk stratification in different LVEF levels and potentially guide decision-making for clinical therapy.

This study aims to identify the risk factors of re-intervention on targeted vessels after balloon pulmonary angioplasty (BPA) in patients with inoperable chronic thromboembolic pulmonary hypertension (CTEPH). We retrospectively analyzed consecutive patients with inoperable CTEPH undergoing BPA at a single center. Targeted vessels and patients were classified according to whether re-intervention occurred. Mixed-effects logistic regression with a patient-level random intercept was used to identify predictors of re-intervention. The optimal threshold was determined by ROC analysis. Time to re-intervention was described by vessel-level Kaplan–Meier curves, and model performance was evaluated using the C-statistic, bootstrap internal validation, calibration plots, and decision-curve analysis. We enrolled 62 patients with inoperable CTEPH who underwent 168 BPA sessions and 412 targeted vessels. Re-intervention occurred in 98/412 vessels (23.8

Accurate prediction of major adverse cardiovascular events (MACEs) is crucial for devising personalized treatments for patients with reperfused myocardial infarction. This requires integrating cardiovascular magnetic resonance (CMR) and electronic health record (EHR) data to enable a comprehensive risk assessment. However, both often suffer from missing modality issues—CMR may lack certain imaging sequences like T2 mapping due to equipment limitations or protocol variations, and EHR data may omit variables that are not measured or recorded. Existing methods address neither these gaps nor the transparent, interpretable reasoning required for trustworthy clinical AI. Here, we propose MACE–MAIS, an end-to-end multimodal AI system with integrated reasoning to predict MACE from incomplete CMR and EHR data. It uses a missing-modality-aware contrastive image pretraining to robustly extract features from incomplete CMR data, and a large language model to embed unstructured EHR texts despite missing entries. Furthermore, MACE–MAIS couples each risk prediction with an interpretable rationale, offering clinicians actionable insights and serving as a reliable reference for routine practice. Evaluated across four real-world clinical datasets, MACE-MAIS outperforms baseline methods in MACE risk prediction while providing transparent, clinically relevant reasoning. This system provides a practical and reliable solution for interpretable clinical AI decision support.

BACKGROUND:MGP (matrix Gla-protein), a known inhibitor of vascular calcification, becomes biologically active by vitamin K-dependent carboxylation. Circulating levels of dpucMGP (dephospho-uncarboxylated matrix Gla-protein), the inactive form of MGP, have been associated with large artery stiffening and reduced skeletal muscle mass in heart failure (HF). Whether dpucMGP is related to adverse outcomes in patients with HF is unknown. METHODS:In this cohort study, we measured plasma dpucMGP among 2247 PHFS (Penn HF Study) participants. We examined the relationship between dpucMGP and ≈5000 other proteins (SomaScan assay) to identify biological pathways associated with dpucMGP. We assessed the association between dpucMGP levels and (1) death or HF-related hospital admission; (2) all-cause death. RESULTS:Participants' median age was 61 years (interquartile range, 53-70 years), 64% were male, and 71% were White. dpucMGP exhibited prominent proteomic associations with acute phase response, coagulation, complement system, fibrosis, cell signaling, and metabolic pathways. Greater dpucMGP was associated with older age, renal dysfunction, and warfarin use, whereas Black ethnicity was associated with lower dpucMGP. Increased dpucMGP levels were associated with an increased risk of death or HF-related hospital admission (standardized hazard ratio, 1.23 [95% CI, 1.17-1.28]; P<0.0001) and all-cause death (standardized hazard ratio, 1.32 [95% CI, 1.25-1.40]; P<0.0001), particularly among participants with nonischemic HF. Associations between dpucMGP and outcomes were dependent on warfarin use, and higher dpucMGP levels were found to mediate the association between warfarin use and adverse outcomes (death [total effect: P=0.005; indirect effect: P<0.001] and death or HF-related hospital admission [total effect: P<0.001; indirect effect: P=0.002]). CONCLUSIONS:Higher dpucMGP is associated with multiple biological pathways and with an increased risk for adverse outcomes in HF. Greater dpucMGP levels mediated the relationship between warfarin use and adverse outcomes. Further studies are required to determine the role of therapeutic interventions to reduce dpucMGP levels in this patient population.

Background Microvascular obstruction (MVO) is strongly associated with adverse outcomes after ST-segment elevation myocardial infarction (STEMI). However, manual quantification of MVO is time-consuming and fails to capture the heterogeneity of microvascular injury. Purpose To evaluate an artificial intelligence (AI)-based model including automated MVO segmentation and radiomic feature extraction to decode microvascular damage heterogeneity, and to assess its ability to predict major adverse cardiovascular events (MACEs). Materials and Methods This multicenter retrospective study (June 2013-December 2023) included patients with STEMI and MVO who underwent cardiac MRI. A previously developed AI model was applied for automated MVO analysis, followed by least absolute shrinkage and selection operator (LASSO) regression for dimensionality reduction to construct a radiomic score (radscore). The primary outcome was MACEs, including cardiovascular death, myocardial reinfarction, malignant arrhythmia, and hospitalization for heart failure. Restricted cubic spline analysis was performed to examine the potentially nonlinear relationship between the radscore and MACE risk. Results Among the 843 patients with STEMI (median age, 60 years [IQR, 51-67 years]; 760 male patients; training set, n = 387; validation set, n = 166; external test set, n = 290), 190 experienced MACEs. The AI model segmented the MRI scans, from which 1595 radiomic features were extracted, and LASSO regression was used to obtain six features for constructing the radscore. Patients with MACEs had higher radscores than those without MACEs (mean, -0.98 ± 0.50 [SD] vs -1.42 ± 0.50; P < .001). Compared with conventional MVO volume quantification, the radscore demonstrated greater prognostic value. The radscore emerged as an independent predictor of MACEs (hazard ratio, 4.20 [95% CI: 3.19, 5.53]; P < .001) and contributed to optimizing risk stratification. Integrating the radscore with conventional variables enhanced prognostic performance, with the C index increasing from 0.77 (95% CI: 0.73, 0.81) for conventional variables alone to 0.80 (95% CI: 0.77, 0.83) for conventional variables plus radscore (P < .001). Conclusion AI-automated MVO radiomic analysis effectively predicted MACE risk and outperformed conventional quantitative assessment. © RSNA, 2026 Supplemental material is available for this article.

To assess the prognostic value of multiple late gadolinium enhancement (LGE) quantification methods for sudden cardiac death (SCD) prediction in hypertrophic cardiomyopathy (HCM). In this single-center retrospective study, 617 HCM patients who underwent cardiac magnetic resonance (CMR) examinations were consecutively enrolled. LGE was quantified using the n-standard deviation (SD) technique (with multiple thresholds) and the full width at half maximum method. “Gray zone” was defined as the myocardium with signal intensity between two predefined thresholds on LGE images. The primary outcome was SCD or aborted SCD. Among 617 HCM patients, 424 (68.7

Artificial intelligence has made significant strides in predicting major adverse cardiovascular events (MACE) in patients with acute myocardial infarction (AMI) following percutaneous coronary intervention. However, most existing methods rely solely on tabular variables derived from clinical data and cardiac magnetic resonance (CMR), without fully leveraging the predictive potential of the CMR imaging modality itself. Moreover, these approaches often overlook the synergistic benefits of multimodal integration between imaging and tabular data. In addition, current models primarily focus on short-term MACE risk assessment (e.g., within 6 months or 1 year), limiting their applicability for long-term prognostication. To address these limitations, we first developed ReconSeg3D, a model that reconstructs short-axis cine CMR stacks into temporally-resolved 3D bi-ventricular volumes, capturing fine-grained cardiac anatomy and dynamic motion. These bi-ventricular sequences were then integrated with 45 clinical and CMR-derived variables using spatiotemporal decomposition and cross-attention mechanisms to construct a multimodal MACE prediction model-HeartTTable. HeartTTable achieved a 5-year time-dependent AUC of 0.934 (95% CI 0.907-0.959) and a Harrell's C-index of 0.897 for predicting MACE risk, significantly outperforming models based solely on clinical and CMR-derived tabular features, and demonstrated strong capabilities in postoperative risk stratification. Our study contributes to improved long-term postoperative management for AMI patients by offering clinicians an objective, data-driven decision-support tool.

Purpose To investigate the association between biventricular imbalance, as reflected by the left ventricular (LV) and right ventricular (RV) global longitudinal strain (GLS) ratio derived from cardiac MRI feature tracking, and all-cause death in isolated left ventricular myocardial infarction. Materials and Methods In this multicenter retrospective study, the primary end point was all-cause mortality. The GLS ratio was defined as the ratio of LVGLS and RVGLS. Long-axis (two, three, and four chambers) and short-axis cine sequences were used for strain analysis. Kaplan-Meier curves, Cox proportional hazards regression, and the C statistic were used for statistical analysis. Results In the internal testing set (919 patients; median age, 59 years [IQR, 51-66 years]; 769 [83.7%] male), 90 (9.8%) patients died over a median follow-up period of 60.9 months (IQR, 47.7-83.6 months). The GLS ratio was the strongest independent predictor for all-cause death (adjusted hazard ratio [HR], 1.69; 95% CI: 1.45, 1.97; P < .001). The model, along with clinical conventional imaging, RVGLS, and GLS ratio, demonstrated improved discrimination (C statistic, 0.82; 95% CI: 0.76, 0.87) and calibration (χ2, 71.88). A GLS ratio of more than 0.95 was responsible for a fourfold death risk after multivariable adjustment. In those with normal RVGLS (HR, 2.43; 95% CI: 1.15, 5.10; P = .02), reserved LVGLS (HR, 5.38; 95% CI: 2.46, 11.79; P < .001), and both (HR, 7.76; 95% CI: 2.16, 11.90; P = .002), a high GLS ratio (>0.95) still predicted all-cause death. Conclusion In isolated LV myocardial infarction, an elevated GLS ratio may provide additional prognostic information for all-cause death.

BACKGROUND:Balloon pulmonary angioplasty (BPA) is an effective therapeutic alternative for patients with chronic thromboembolic pulmonary hypertension (CTEPH), which improved pulmonary arterial compliance (CPA) and pulmonary vascular resistance (PVR). OBJECTIVE:To investigate whether the CPA is a predictor of exercise tolerance after BPA. METHODS:The correlations between changes in each parameter and changes in six-minute walking distance (6MWD) were evaluated by Pearson's test. The determinants of functional capacity that was defined as 6MWD ≥440 m were assessed with a logistic regression model. Multiple linear regression analysis was used to identify the independent variables related to △6MWD. RESULTS:We enrolled 70 patients (female/male: 40/30, mean age: 64 years) who underwent a total of 271 BPA sessions which significantly increased CPA [1.0 (0.7, 1.3) vs. 2.1 (1.7, 2.5) mL/mmHg], and decreased PVR [6.7 (3.6, 9.7) vs. 3.0 (2.1, 4.3) wood units]. The correlation coefficient between improvement in 6MWD and changes in CPA was r = 0.328 (P = 0.006). At univariate analysis, duration of pulmonary hypertension symptoms and pulmonary arterial compliance were found to be associated with good exercise tolerance. Multivariate analysis demonstrated that CPA (95 %CI: 1.23 to 3.75, P = 0.026) was an independent predictor of exercise tolerance after BPA. The multiple linear regression analysis demonstrated that △CPA (β= 0.292, P = 0.019) was an independent predictor of △6MWD. CONCLUSION:BPA significantly improved CPA in inoperable patients with CTEPH and the resistance-compliance relationship maintained inversely associated. After successful BPA, baseline CPA is an important determinant of exercise tolerance.

Background Current risk stratification models fail to effectively integrate a broad range of parameters to predict major adverse cardiovascular events (MACE) in patients with ST-segment elevation myocardial infarction (STEMI). Purpose To develop and externally test a machine learning (ML) model integrating comprehensive clinical data and cardiac MRI parameters for long-term MACE prediction in patients with STEMI. Materials and Methods This retrospective study included data from patients with STEMI who underwent clinically indicated cardiac MRI within 7 days after percutaneous coronary intervention, with data from one center composing the training set (September 2015 to September 2023) and data from another center composing the external test set (January 2015 to July 2023). The primary end point was MACE, defined as a composite of cardiovascular death, recurrent myocardial infarction, unplanned coronary revascularization, stroke, and rehospitalization for heart failure or arrhythmia. Sixty-seven variables were initially evaluated to inform the ML model. The final model included established clinical predictors combined with features selected using recursive feature elimination. Model performance was assessed using integrated area under the receiver operating characteristic curve (AUC). Results A total of 1066 patients with STEMI (mean age, 58.15 years ± 11.40 [SD]; 904 male patients) were included: 682 in the training set and 384 in the external test set. During a median follow-up of 40 months (IQR, 22-55 months), 142 patients in the training set and 81 in the external test set experienced MACE. In the external test set, the ML model achieved an integrated AUC for MACE prediction of 0.91, compared with an integrated AUC of 0.86 for a clinical model (P < .001), 0.86 and 0.89 (P = .005) for Cox regression models, 0.66 for Global Registry of Acute Coronary Events score (P < .001), and 0.62 for Thrombolysis in Myocardial Infarction score. The model effectively stratified patients into distinct risk groups (log-rank P < .001). Conclusion An ML model integrating cardiac MRI and clinical data demonstrated excellent long-term prognostic performance compared with traditional models and aided individualized risk stratification in patients with STEMI. © RSNA, 2026 Supplemental material is available for this article. See also the editorial by Garot and Duhamel in this issue.

BACKGROUND:Heterogeneity of heart failure with preserved ejection fraction (HFpEF) results in significant challenges for treatment development. Identifying and characterising distinct HFpEF phenogroups may aid in tailoring therapeutic strategies for these patients. The objective of this study was to assess proteomic patterns of HFpEF phenogroups identified through a machine-learning-based clustering model, with the aim of uncovering specific biological pathways associated with each phenogroup. METHODS:This study represents a post-hoc analysis of the ongoing Prospective mUlticenteR obServational stUdy of patIenTs with Heart Failure with preserved Ejection Fraction (PURSUIT-HFpEF) study, which is a multicentre prospective observational study of hospitalised patients with acute decompensated HFpEF. Of the overall cohort (N=1238), this study analysed 198 patients with HFpEF with available proteomics data. These patients were classified into four phenogroups using the machine-learning-based clustering model. The SomaScan assay V.4.1 was used to measure levels of >7000 plasma proteins, and subsequent pathway analysis was conducted to determine the biological differences among the phenogroups. RESULTS:We identified four distinct phenogroups: Phenogroup 1 ('rhythm trouble'), Phenogroup 2 ('ventricular-arterial uncoupling'), Phenogroup 3 ('low output and systemic congestion') and Phenogroup 4 ('systemic failure'). The proteomics revealed distinct protein expression profiles among the phenogroups, with ribonuclease 4, tax1-binding protein 1, regenerating islet-derived protein 3-gamma and alpha-1-antichymotrypsin being the most significant markers to specific identified phenogroups. Pathway analysis suggested differences in immune response, autonomic activation, cellular homeostasis and tissue repair mechanisms across the phenogroups. CONCLUSIONS:Using a comprehensive plasma proteomics approach, our study identified distinct proteomic profiles of HFpEF phenogroups, which in turn suggest specific underlying biological processes. These profiles suggest the involvement of inflammatory activation, tissue injury and regenerative responses, immune modulation and systemic stress signalling as key components of HFpEF pathophysiology. TRIAL REGISTRATION NUMBER:UMIN-CTR ID: UMIN000021831.

Transglutaminase 2 (TGM2) has been implicated in various health conditions, yet its role in acute coronary syndrome (ACS) remains poorly characterized in clinical settings. This study investigated the association between circulating TGM2 levels and the severity of coronary stenosis in ACS. A total of 242 individuals with ACS were included in this study. Clinical data were collected, and the severity of coronary stenosis was evaluated with the Gensini and Syntax scoring systems. Kaplan-Meier analysis, logistic regression, and receiver operating characteristic (ROC) curve analysis were performed. Circulating TGM2 levels were significantly higher in the STEMI group (176.3 pg/mL) and the non-STEMI group (181 pg/mL) than the UA group (64 pg/mL) and the stable CAD group (50.95 pg/mL) (P < 0.001). Multivariate analysis, after adjustment for confounding factors, identified TGM2 as an independent risk factor for acute myocardial infarction (odds ratio: 44.292 per 100 pg/mL increase in TGM2; 95% CI: 2.491–7.398; P < 0.001). During a median follow-up of 477 days, Kaplan-Meier survival analysis demonstrated that patients with higher TGM2 levels (≥91.9 pg/mL) exhibited a significantly lower MACE-free survival rate (P = 0.0142). ROC curve analysis further revealed that combining TGM2 and Gensini scores yielded superior predictive performance for MACE to that of either parameter alone. Circulating TGM2 is elevated in ACS and is strongly associated with the presence of AMI. Furthermore, it provides prognostic information for MACE, particularly when it is used in combination with established anatomical risk scores.

BACKGROUND:MRI is important for cardiac disease evaluation, but accurate diagnosis remains challenging in less experienced centers. Although large language models (LLMs) have shown promise in medical imaging diagnosis, their application in cardiac MRI is limited. HYPOTHESIS:LLMs may be effective in achieving cardiac MRI diagnosis based on standardized descriptions. STUDY TYPE:Retrospective. POPULATION:A total of 203 hypertrophic cardiomyopathy, 186 dilated cardiomyopathy, 46 hypertensive heart disease, 198 ischemic cardiomyopathy, 38 constrictive pericarditis, 45 cardiac amyloidosis, 91 myocarditis, and 144 normal controls. FIELD STRENGTH/SEQUENCES:Balanced steady-state free-precession, short tau inversion recovery, and breath-hold inversion-recovery segmented gradient-echo sequences at 3.0 T. ASSESSMENT:Clinical and cardiac MRI information from each subject was converted into standardized descriptions and input into Generative Pre-trained Transformer-4.5 (GPT-4.5), GPT-4 Omni (GPT-4o), Deepseek-V3, and Deepseek-R1 LLMs. Cardiac MRI information included LV function, wall thickness and motion, and abnormalities on T2WI, perfusion and late gadolinium enhancement sequences. Each model was asked to generate an imaging diagnosis. In addition, a medical student (8 months experience) and three radiologists (junior, mid-level and senior: with 3, 6, and 10 years' experience, respectively) provided diagnoses based on cardiac MRI images and clinical information. STATISTIC TESTS:Frequency-weighted sensitivity and specificity were calculated. The diagnostic performances of the LLMs and human readers were compared using the McNemar test with Bonferroni correction. A p value < 0.05 was considered significant. RESULTS:All LLMs showed excellent frequency-weighted specificity (0.973-0.983). The frequency-weighted sensitivities of all LLMs were not significantly different from that of the junior radiologist, were significantly higher than that of the medical student, and significantly inferior to those of the senior radiologist (GPT-4.5: 0.863, GPT-4o: 0.821, Deepseek-V3: 0.843, and Deepseek-R1: 0.851 vs. junior radiologist: 0.850, all adjusted p = 1.000; vs. medical student: 0.731, all adjusted p < 0.001; vs. senior radiologist: 0.942, all adjusted p < 0.001). Additionally, the mid-level radiologist achieved a frequency-weighted sensitivity of 0.895, outperforming all LLMs except GPT-4.5. DATA CONCLUSION:LLMs may generate accurate diagnoses from standardized cardiac MRI descriptions, potentially benefiting less experienced physicians. LEVEL OF EVIDENCE: 4: TECHNICAL EFFICACY:Stage 5.

Background Risk stratification in hypertrophic cardiomyopathy (HCM) is essential for preventing adverse outcomes. The left atrioventricular coupling index (LACI) reflects atrial–ventricular interaction, but its prognostic value in HCM remains unclear. Purpose This study aims to evaluate the incremental prognostic value of LACI in predicting sudden cardiac death (SCD) events in patients with HCM. Materials and Methods In this retrospective study, 2,240 HCM patients were analyzed. Clinical and CMR parameters, including LACI, late gadolinium enhancement (LGE), and left ventricular ejection fraction (LVEF), were evaluated. LACI was defined as the ratio of left atrial end-diastolic volume to left ventricular end-diastolic volume. Univariable and multivariable Cox regression analyses assessed associations of LACI with SCD events and guideline-based models (ESC and ACC/AHA). Results Over a mean follow-up of 4.0 ± 2.5 years, 128 patients (5.7%) experienced SCD events. Event patients had higher LACI (39% ± 28% vs. 33% ± 23%, P = 0.04), greater LGE (12.1 ± 8.6% vs. 6.4 ± 4.3%, P < 0.001), and lower LVEF (48 ± 15% vs. 65 ± 13%, P < 0.001). LACI independently predicted SCD events after adjusting for LGE and conventional risk factors (multivariable HR 1.44–1.46, 95% CI 1.3–1.6; P < 0.001) in both ESC- and ACC/AHA-adjusted models. Adding LACI and LGE to the ESC model improved AUC (0.69 vs. 0.53) and specificity (82.6% vs. 57.9%). Conclusion CMR-derived LACI independently predicts adverse outcomes in HCM and provides incremental prognostic value beyond LGE and guideline-based models.

RATIONALE AND OBJECTIVES:To evaluate the prognostic value of radiomic features derived from contrast-free cine cardiac magnetic resonance (CMR) in patients with ST-segment elevation myocardial infarction (STEMI). MATERIALS AND METHODS:We retrospectively included 440 patients with acute STEMI (86.6% males, 56.9 ± 10.6 years of age), who underwent CMR one week after percutaneous coronary intervention. Patients were assigned by centers into a development cohort (n = 359) and a validation cohort (n = 81). Radiomic features were extracted from cine images. Feature selection was performed using random survival forest and least absolute shrinkage and selection operator (LASSO)-Cox regression to generate a radiomics-based risk score (RAD score). Discrimination was evaluated using logistic and Cox regression analysis. RESULTS:During the median follow-up period of 2.9 years, 88 patients experienced major adverse cardiovascular events (MACE). The RAD score provided incremental prognostic value over the clinical model in the internal (C-index 0.86 [0.79-0.92] vs 0.65 [0.55-0.78]; p < 0.001) and external cohort (C-index 0.80 [0.70-0.91] vs 0.63 [0.48-0.78]; p = 0.014), comparable to the clinical + LGE-CMR model (C-index 0.80 [0.70-0.91] vs 0.77 [0.65-0.89]; p = 0.547). Receiver operating characteristic analyses were consistent with C-index findings. After adjusting for established risk factors, RAD score-defined high risk remained independently associated with MACE (HR 11.30, 95% CI 4.96-21.44; p < 0.001). CONCLUSION:Cine-CMR radiomics provides independent and incremental prognostic information after STEMI and attains predictive performance comparable to parameters from cardiac magnetic resonance with late gadolinium enhancement, supporting contrast-free, individualized risk stratification.

We derived and validated proteomic risk scores (PRSs) for heart failure (HF) prognosis that provide absolute risk estimates for all-cause mortality within 1 year. Plasma samples from individuals with HF with reduced ejection fraction (HFrEF; ejection fraction <40%; training/validation n=1247/762) and preserved ejection fraction (HFpEF; ejection fraction ≥50%; training/validation n=725/785) from 3 independent studies were run on the SomaScan Assay measuring ≈5000 proteins. Machine learning techniques resulted in unique 17- and 14-protein models for HFrEF and HFpEF that predict 1-year mortality. Discrimination was assessed via C-index and 1-year area under the curve (AUC), and survival curves were visualized. PRSs were also compared with Meta-Analysis Global Group in Chronic HF (MAGGIC) score and NT-proBNP (N-terminal pro-B-type natriuretic peptide) measurements and further assessed for sensitivity to disease progression in longitudinal samples (HFrEF: n=396; 1107 samples; HFpEF: n=175; 350 samples). In validation, the HFpEF PRS performed significantly better (P≤0.1) for mortality prediction (C-index, 0.79; AUC, 0.82) than MAGGIC (C-index, 0.71; AUC, 0.74) and NT-proBNP (PRS C-index, 0.76 and AUC, 0.81 versus NT-proBNP C-index, 0.72 and AUC, 0.76). The HFrEF PRS performed comparably to MAGGIC (PRS C-index, 0.76 and AUC, 0.83 versus MAGGIC C-index, 0.75 and AUC, 0.84) but had a significantly better C-Index (P=0.026) than NT-proBNP (PRS C-index, 0.75 and AUC, 0.78 versus NT-proBNP C-index, 0.73 and AUC, 0.77). PRS included known HF pathophysiology biomarkers (93%) and novel proteins (7%). Longitudinal assessment revealed that HFrEF and HFpEF PRSs were higher and increased more over time in individuals who experienced a fatal event during follow-up. PRSs can provide valid, accurate, and dynamic prognostic estimates for patients with HF. This approach has the potential to improve longitudinal monitoring of patients and facilitate personalized care.

BACKGROUND:Microvascular occlusion (MVO) determined by cardiac magnetic resonance (CMR) exists both in acute phase and recovery period after myocardial infarction. This study aimed to examine the long-term prognosis predictive value of persistent MVO for ST-segment elevation myocardial infarction (STEMI). METHODS:A prospective cohort enrolled 344 patients with STEMI who received primary percutaneous coronary intervention and underwent CMR both in 5 to 7 days and 6 months after STEMI to determine if MVO had occurred. All patients were followed up for 5 years, and major adverse cardiovascular events (MACE) were recorded. RESULTS:This study included 344 STEMI patients with an average age of 57 years at 6 centers in China. A total of 192 (55.81%) patients with STEMI did not have MVO by CMR, and 105 (30.52%) patients showed transient MVO in acute phase of myocardial infarction and 47 (13.66%) patients showed persistent MVO at 6 months after infarction. The patients with persistent MVO had the largest infarct size and the lowest left ventricular ejection fraction both in 5 to 7 days and 6 months after infarction (all P < .001). Patients with persistent MVO showed a significantly higher incidence of 5-years MACE than those without MVO or with MVO in only 1 week (66.0% vs 18.8% and 27.6%, respectively; P < .001). Persistent MVO was an independent strong predictor of MACE after adjustment for other CMR variables (OR: 3.912, 95% CI: 1.904-8.037; P < .001). A propensity score-matched population comprised 43 patients with persistent MVO and 43 patients with transient MVO in only 1 week. The patients with persistent MVO had a higher incidence of MACE than those with transient MVO (65.1% [28/43] vs 37.2% [16/43]; P = .010). CONCLUSION:Persistent MVO by CMR at the chronic phase of STEMI provides useful prognostic information regarding long-term outcomes after primary percutaneous coronary intervention.

BACKGROUND: Iron deficiency (ID) is currently defined as a serum ferritin level <100 or 100 to 299 ng/mL with transferrin saturation (TSAT) <20%. Serum ferritin and TSAT are currently used to define absolute and functional ID. However, individual markers of iron metabolism may be more informative than current arbitrary definitions of ID. METHODS: We assessed prognostic associations of ferritin, serum iron, and TSAT among 2050 participants with heart failure (HF) with reduced/mid-range (n=1821) or preserved (n=229) left ventricular ejection fraction enrolled in the PHFS (Penn HF Study), a prospective cohort study. We measured 4928 plasma proteins using an aptamer-based assay (SOMAScanv4) and assessed prognostic and proteomic associations of markers of iron metabolism. RESULTS: Ferritin concentrations were not associated with outcomes, whereas low TSAT and serum iron were associated with the risk of all-cause death (TSAT: standardized hazard ratio, 0.84 [95% CI, 0.76–0.93]; P =0.001; serum iron: standardized hazard ratio, 0.87 [95% CI, 0.79–0.96]; P =0.007). Similarly, TSAT was associated with the risk of death or HF-related admission (standardized hazard ratio, 0.89 [95% CI, 0.83–0.95]; P =0.0006). Significant interactions between TSAT and HF with preserved ejection fraction status were found such that TSAT was more strongly associated with the risk of death and death or HF-related admission in HF with preserved ejection fraction. We identified 359 proteins associated with TSAT, including TFRC (transferrin receptor protein; β, −0.455; P <0.0001) and CRP (C-reactive protein; β, −0.355; P <0.0001). Pathway analyses demonstrated associations with lipid metabolism, complement activation, and inflammation. In contrast to the robust associations between TSAT and outcomes, ID and absolute ID defined by current criteria were not associated with death or death or HF-related admission. TSAT was associated with outcomes regardless of the presence of functional versus absolute ID. CONCLUSIONS: Low TSAT, but not ferritin concentrations, is significantly associated with adverse outcomes in HF. Low TSAT is more strongly associated with outcomes in HF with preserved ejection fraction. Pathways related to inflammation and lipid metabolism are associated with low TSAT in HF.

BACKGROUND:Microvascular obstruction (MVO) is associated with heart failure (HF) following ST-segment-elevation myocardial infarction. Angiography-derived microcirculatory resistance (AMR), a wire- and adenosine-free measure, may facilitate early assessment of microvascular function post-primary percutaneous coronary intervention. This study aimed to evaluate the ability of AMR to detect MVO and its prognostic value for predicting HF in patients with ST-segment-elevation myocardial infarction post-primary percutaneous coronary intervention. METHODS:Patients with consecutive ST-segment-elevation myocardial infarction undergoing primary percutaneous coronary intervention with a cardiac magnetic resonance examination 2 to 7 days post-procedure between April 2016 and February 2023 were retrospectively reviewed. AMR was computed from coronary angiography. MVO was identified and quantified via cardiac magnetic resonance. The end point was new-onset HF during follow-up. RESULTS:Overall, 475 patients (aged 56.8±11.7 years; 399 men) were included. The area under the curve for AMR to detect MVO was 0.821 (95% CI, 0.782-0.859), with an optimal cutoff value of 2.7 mm Hg*s/cm. During a median follow-up of 37.3 months, 121 (25.5%) patients developed HF. AMR, whether as a continuous (per 0.5-mm Hg*s/cm increase; hazard ratio, 1.29 [95% CI, 1.10-1.52]; P=0.002) or categorical (AMR >2.7 mm Hg*s/cm; hazard ratio, 2.15 [95% CI, 1.43-3.22]; P<0.001) variable, was independently associated with HF after adjusting for traditional risk factors (age, symptom-to-balloon time, left anterior descending coronary artery, and ejection fraction) and late gadolinium enhancement-cardiac magnetic resonance parameters. AMR improved prognostication over traditional risk factors and late gadolinium enhancement-cardiac magnetic resonance parameters (net reclassification improvement, 0.533; P<0.001; integrative discrimination index, 0.023; P=0.005). CONCLUSIONS:AMR showed good diagnostic performance in detecting MVO and was an independent and incremental predictor of HF in patients with ST-segment-elevation myocardial infarction post-primary percutaneous coronary intervention.