Cardiovascular risk factors (RFs) are commonly used in clinical practice to predict future adverse cardiovascular outcomes, including acute coronary syndromes (ACS). A recent study reported an association between RFs and plaque vulnerability in patients with ACS. This study aimed to investigate the association between RFs (modifiable and non-modifiable) and pan-coronary plaque burden, plaque phenotype, and features of vulnerability. In patients undergoing 3-vessel optical coherence tomography imaging, modifiable (dyslipidemia, hypertension, diabetes mellitus, obesity, smoking) and non-modifiable (age, sex, family history) RFs were recorded. Plaque number, plaque phenotype, and vulnerable features were analyzed. A total of 534 plaques from 131 patients (36.6% ACS) were analyzed. As the number of RFs increased, the number of plaques (P trend = 0.001) as well as plaques with a vulnerable phenotype (thin-cap fibroatheromas [TCFAs], P trend = 0.001) increased. An increasing number of RFs was also associated with a higher number of vulnerable features (P trend < 0.001), including more thin fibrous caps (P trend = 0.001), lipid-rich plaques (LRPs), macrophages, microvessels, and cholesterol crystals (all P trend ≤ 0.001). In multivariable analyses, only modifiable RF burden was associated with increased pan-coronary vulnerability (incidence rate ratio [IRR]: 1.36; 95% CI: 1.18-1.57; P < 0.001), including a higher prevalence of TCFAs (IRR: 1.48; 95% CI: 1.20-1.83; P < 0.001), LRPs (IRR: 1.35; 95% CI: 1.15-1.59; P < 0.001), and cholesterol crystals (IRR: 1.50; 95% CI: 1.21-1.87; P < 0.001). As the number of cardiovascular RFs increased, the number of plaques, plaques with a high-risk phenotype, and vulnerable features also increased. Only increasing modifiable RF burden was associated with greater pan-coronary vulnerability, including more TCFAs and LRPs. ClinicalTrials.gov NCT01110538.
Right ventricular (RV) involvement is a potential high-risk marker in hypertrophic cardiomyopathy (HCM); however, its prognostic relevance in apical HCM remains unclear. This study aimed to evaluate the prognostic significance of RV involvement in apical HCM and the incremental value of imaging-based phenotyping. We retrospectively analyzed 289 patients with apical HCM who underwent cardiovascular magnetic resonance and strain echocardiography. Patients were classified into 3 morphological subtypes-purely apical, mixed, and RV involvement. The primary outcome was a composite of cardiovascular (CV) events. In total, 117 patients (40.5%) had purely apical HCM, 102 (35.3%) had mixed apical HCM, and 70 (24.2%) had RV involvement. During a median follow-up of 7.7 years, 35 (12.1%) patients experienced CV events. RV involvement was associated with a higher risk of CV events (adjusted HR: 2.59; 95% CI: 1.02-6.60; P = 0.046). Imaging-based clustering identified 2 distinct phenotypic subgroups: Cluster 1 (n = 200) and cluster 2 (n = 89). Cluster 2 was characterized by greater hypertrophy, more fibrosis, and impaired systolic/diastolic function, and was associated with a higher risk of CV events (adjusted HR: 2.31; 95% CI: 1.04-5.13; P = 0.039). When stratified into 4 groups according to cluster and RV involvement, patients in cluster 2 with RV involvement exhibited the highest risk (adjusted HR: 4.02; 95% CI: 1.44-11.19; P = 0.008). RV involvement is associated with adverse prognosis in apical HCM, and imaging-based clustering further identifies high-risk phenotypes. Combined risk stratification incorporating imaging-derived clusters and RV morphology may enhance individualized risk assessment in this population.
Heart failure with preserved ejection fraction (HFpEF) is a clinical syndrome characterized by dyspnea caused by hemodynamic congestion, which develops at rest or with exercise. Noninvasive diagnosis remains challenging, because resting diastolic pressures may be normal, and even when abnormal, currently applied noninvasive indices lack sensitivity. The pathophysiology of HFpEF extends far beyond diastolic dysfunction, and such complexity can be better understood through imaging. This State-of-the-Art Review emphasizes echocardiography and other noninvasive imaging approaches as the first-line tools for HFpEF evaluation, while linking multimodality imaging to invasive hemodynamics as a physiologic reference standard. The authors emphasize evidence-based, multiparametric assessment, where combined cardiopulmonary exercise testing with exercise stress echocardiography can quantify the diverse reserve limitations that characterize HFpEF, including cardiac and extracardiac contributors, atrial and right heart dysfunction, pulmonary vascular and ventilatory limitation, dynamic atrioventricular regurgitation, extrinsic restraint, pulmonary congestion, and peripheral impairments, acknowledging that such phenotyping must be anchored to actionable findings and evidence-based interventions. The authors describe the potential role of exercise cardiac magnetic resonance and artificial intelligence-enabled imaging analysis for phenotyping and risk stratification. Finally, the authors propose an echo-first Bayesian pathway that updates pretest to posttest probability and supports identification of patients in whom invasive hemodynamic exercise testing may be required to confirm the diagnosis and validate imaging markers against invasively measured congestion.
Marfan syndrome (MFS) is a heritable connective tissue disorder in which cardiovascular complications drive morbidity. Contemporary sex-based differences in vascular and rhythm manifestations remain incompletely defined. The purpose of this study was to compare aneurysm burden, dissection prevalence, arrhythmias, valvular phenotypes, and cardiovascular interventions between men and women with confirmed MFS. We performed a retrospective cohort study of adults with confirmed MFS evaluated from 2018 to 2024 at Mayo Clinic. Diagnoses and vascular outcomes were confirmed by manual chart and imaging report review. Outcomes reflected ever-documented history during the study period. Multivariable logistic regression estimated sex-based associations adjusted for age, hypertension, diabetes mellitus, chronic kidney disease, and heart failure. Sensitivity analyses were performed in the genetically positive subgroup. Among 783 patients (443 men, 340 women), men more often had ascending aortic aneurysm (89.1% vs 71.1%), any aneurysm (91.0% vs 75.3%), and any extra-aortic aneurysm (36.8% vs 25.9%) (all P ≤ 0.001). Dissection prevalence was similar by sex. Men also had more atrial fibrillation (44.0% vs 30.9%), atrial flutter (18.3% vs 12.1%), sustained ventricular tachycardia (9.7% vs 5.0%), and aortic root surgery (63.4% vs 40.3%) (all P < 0.05), whereas women more often had mitral valve prolapse (45.6% vs 37.7%; P = 0.028). After adjustment, male sex remained associated with aneurysmal burden, atrial fibrillation, sustained ventricular tachycardia, and aortic root surgery. In this multicenter Marfan cohort, men had a greater burden of aneurysmal disease, atrial arrhythmias, sustained ventricular tachycardia, and aortic intervention, whereas dissection prevalence was similar between sexes.
High lipoprotein(a) (Lp[a]) associates with elevated coronary heart disease event risk, but it remains uncertain whether this excess risk can be explained by increased coronary plaque burden. This study aims to determine whether Lp(a)-related excess cardiovascular risk in the general population is primarily explained by increased coronary atherosclerosis or by additional plaque vulnerability-related effects. This study included 28,529 middle-aged individuals from the SCAPIS (Swedish CArdioPulmonary bioImage Study), a population-based cohort. Coronary atherosclerosis was assessed using coronary computed tomography angiography in 25,916 participants. Incident coronary events, defined as fatal or nonfatal myocardial infarction, were ascertained over a median follow-up of 7.8 years. High Lp(a) was present in 14% of participants and associated with severe/extensive coronary atherosclerosis (Segment Involvement Score >4; OR: 1.38; 95% CI: 1.21-1.57) and coronary events (HR: 1.54; 95% CI: 1.18-2.0). Stratified, adjusted, and mediation analyses consistently suggested that the association between Lp(a) and coronary event risk is largely explained by greater atherosclerotic burden (P for mediation <0.0001). Individuals with high Lp(a) developed atherosclerosis 2 to 3 years earlier and the population-attributable fraction of Lp(a) was 5% to 10% for severe atherosclerosis and around 7% for coronary events. The findings suggest that Lp(a) increases coronary event risk primarily by driving a higher burden of atherosclerosis and that high Lp(a) accounts for 5% to 10% of total coronary disease burden in the population. These results support broad Lp(a) measurements and the use of coronary imaging for risk stratification in individuals with high Lp(a).
Metabolic and musculoskeletal abnormalities in patients with heart failure (HF) may not be fully captured by body mass index. Body composition was studied in participants with HF and matched control subjects in the UK Biobank Imaging Study, and findings were compared among those with HF with reduced ejection fraction, HF with mildly reduced ejection fraction, and HF with preserved ejection fraction. Participants with HF and available ejection fraction (n = 185) were matched 1:5 to age-, sex-, and body mass index-matched control subjects (n = 925). Whole-body magnetic resonance-quantified visceral adipose tissue (VAT), abdominal subcutaneous adipose tissue, liver fat, thigh fat-free muscle volume (MV), and muscle fat infiltration (MFI) were assessed. Personalized z-scores adjusted for sex and body size were derived. Adverse muscle composition (AMC) was defined as low MV z-score and high MFI. Dual-energy x-ray absorptiometry was used to measure total fat and appendicular lean mass. Compared with control subjects, HF participants had higher VAT (z-score = 0.46 ± 1.1 vs 0.07 ± 1.0; P < 0.001), MFI (8.4 ± 2.3% vs 7.6 ± 2.0%; P < 0.001), and AMC (36% vs 17%; P < 0.001) and lower MV (z-score = -0.7 ± 1.0 vs -0.1 ± 0.9; P < 0.001). Despite the muscle derangements identified by magnetic resonance imaging assessment, the prevalence of sarcopenia diagnosed using combined grip strength and dual-energy x-ray absorptiometry-derived lean mass parameters varied considerably depending on the criteria applied. Among HF subtypes, HF with mildly reduced ejection fraction showed the highest VAT and lowest MV, HF with preserved ejection fraction had greater fat and MFI, and HF with reduced ejection fraction had the greatest prevalence of AMC and the weakest grip strength. Participants with HF showed higher VAT and adverse muscle changes, with weaker grip strength. Sarcopenia prevalence varied substantially across definitions, reflecting substantial variability and discrepancies among current sarcopenia definitions and criteria. Traditional anthropometric indexes underestimate the burden of body composition derangements in HF.
暂无摘要(点击查看详情)
[Figure: see text]
暂无摘要(点击查看详情)
In patients with heart failure (HF) with reduced ejection fraction (HFrEF) from acquired heart disease, guideline-directed medical therapy (GDMT) for HF is associated with improved left ventricular (LV) ejection fraction (LVEF) (ie, HF with improved EF (HFimpEF), and which in turn is associated with improved survival. Similar data are lacking in patients with congenital heart disease (CHD). The objective of the study was to describe the prevalence, correlates, and prognostic implications of HFimpEF in adults with CHD, biventricular physiology, and systemic LV. Retrospective study of adults with CHD, biventricular physiology, and systemic LV presenting with HFrEF (2003-2023). Echocardiogram was performed at baseline and 1-year follow-up encounters. GDMT use was assessed at baseline and 1-year follow-up encounters using GDMT score. GDMT uptitration (ΔGDMT) was calculated as the difference between GDMT scores. HFimpEF was defined as HF with baseline LVEF ≤40%, and a subsequent absolute LVEF increase ≥10%, leading to LVEF >40% at follow-up echocardiogram. HFrEF relapse was defined as decline in absolute LVEF >10% leading to LVEF ≤40% in patients with HFimpEF. Cox regression analysis was used to assess the relationship between HFimpEF, GDMT score, and outcomes (death and cardiovascular events). Of the 327 patients (age 46 ± 16 years; LVEF 31% ± 6%), 63 (19%) had HFimpEF. GDMT use (higher baseline and ΔGDMT) was associated with great odds of HFimpEF. HFimpEF was associated with a 26% decrease in all-cause mortality, and 28% decrease in cardiovascular events on multivariable analysis compared to patients without improvement in LVEF. Of the 63 patients with HFimpEF, 27% had HFrEF relapse, and HFrEF relapse was associated with higher risk of cardiovascular events. These data support the use and optimization of GDMT in this subgroup of CHD patients, and the need for ongoing clinical and imaging surveillance.
Wild-type transthyretin amyloid cardiomyopathy (wtATTR-CM) is increasingly recognized with advances in noninvasive imaging and disease-modifying therapies. However, atypical imaging findings may mimic other cardiomyopathies and delay diagnosis. An 85-year-old man was treated for presumed isolated cardiac sarcoidosis after 18F-fluorodeoxyglucose (FDG)/13N-ammonia positron emission tomography/computed tomography (PET/CT) showed matching perfusion defects with focal FDG uptake. Despite methotrexate therapy, he developed progressive heart failure. Serial echocardiography demonstrated increasing left ventricular wall thickness and worsening global longitudinal strain with apical sparing, raising suspicion for amyloidosis. Repeat PET/CT showed persistent perfusion defects but resolution of FDG uptake. Cardiac amyloid scintigraphy revealed grade 3 myocardial tracer uptake (heart-to-contralateral ratio 1.74). Negative monoclonal studies and transthyretin genetic testing confirmed wtATTR-CM. Methotrexate was discontinued, tafamidis was initiated, and symptoms improved. In wtATTR-CM, 18F-FDG/13N-ammonia PET/CT may demonstrate myocardial FDG uptake, mimicking inflammatory cardiomyopathy such as cardiac sarcoidosis. Proposed mechanisms include localized inflammation from amyloid deposition, altered myocardial glucose metabolism, or coexistence of inflammatory processes. Serial multimodality imaging was essential for diagnostic clarification and timely targeted therapy. wtATTR-CM may present with myocardial FDG uptake, mimicking inflammatory cardiomyopathy. Multimodality imaging is critical to avoid diagnostic delay and guide appropriate treatment.
Anderson-Fabry disease (AFD) is a rare cause of left ventricular hypertrophy (LVH) that can be challenging to diagnose, particularly when comorbidities are present. We report a 70-year-old man with controlled hypertension, diabetes mellitus, chronic kidney disease, prior percutaneous coronary intervention on the left anterior descending artery, evaluated for progressive LVH and electrocardiogram (ECG) changes during follow-up. ECG showed an atypical pattern with right bundle branch block. Cardiac magnetic resonance then demonstrated reduced native T1 values. Genetic testing identified the p.N215S (p.Asn215Ser) GLA mutation, confirming late-onset AFD. Disease-specific management was subsequently initiated. Late-onset AFD is frequently underdiagnosed, particularly in older patients with multiple cardiovascular comorbidities. Careful ECG analysis may provide key diagnostic clues and prompt advanced imaging and genetic testing. AFD should be considered in older patients with unexplained LVH. Careful evaluation of ECG abnormalities may be sufficient to trigger targeted, second-level, diagnostic tests.
Rapid diagnostic screening is an unmet need in patients with suspected cardiomyopathy (CM). The authors recently demonstrated in a retrospective study that magnetocardiography (MCG) may be a suitable tool to detect CM. This study aims to prospectively evaluate the accuracy of MCG as a screening tool in patients with suspected CM. MCG uses a superconducting quantum interference device to detect the magnetic field of the heart. In another recent retrospective study by the same investigators, a T-beg-Tmax (MCG score) ≥0.051 identified patients with CM (dilated, hypertrophic, nondilated left ventricular, and restrictive CM) after exclusion of coronary artery disease. The authors assessed the diagnostic accuracy of MCG compared with advanced imaging and/or endomyocardial biopsy. The authors enrolled 110 adult patients with angina-like symptoms after exclusion of coronary artery disease from the emergency department, inpatient wards, and outpatient clinics. A total of 220 healthy individuals were incorporated into the study as control subjects by using 1:2 direct matching (age decade and sex) for a total of 330 participants. MCG detected CM with 94.74% sensitivity (95% CI: 85%-99%), 98.54% specificity (95% CI: 96%-100%), 93.1% negative predictive value (95% CI: 84%-98%), and 94.34% positive predictive value (95% CI: 84%-98%). In 3 patients, cardiac magnetic resonance failed to detect CM, whereas MCG showed pathologic results consistent with results of endomyocardial biopsy. Results were highly reproducible (statistical uncertainty: 2.5%, T-beg-Tmax [interval and electrical vector from the beginning of a T-wave to its peak (MCG score)] ±0.004). MCG offers a noninvasive, reproducible methodology for rapid and accurate detection of CM after exclusion of ischemic heart disease. Our data indicate that MCG may be an efficient tool for allocation of advanced imaging to those patients with a high likelihood of disease. (Magnetocardiography as a Diagnostic Screening Tool for Myocarditis and Other Types of Cardiomyopathy [MagMa]; NCT06689098).
Transradial access is the preferred approach for coronary interventions, but vascular complications may rarely lead to unexpected coronary events. A 50-year-old woman with cardiovascular risk factors presented with exertional symptoms and a positive stress echocardiography result in the anterior territory. Coronary angiography via right radial access revealed a significant mid-left anterior descending coronary artery lesion. During ad hoc percutaneous coronary intervention (PCI), difficult catheter advancement due to radial resistance and spasm required escalation to a long guide sheath. After guidewire placement, angiography revealed a new hazy filling defect at the proximal left anterior descending coronary artery-left circumflex coronary artery bifurcation, associated with chest pain and ST-segment elevation. Thromboaspiration restored coronary flow, retrieving a 3-cm vascular fragment. Histopathology confirmed an avulsed radial arterial wall fragment containing the 3 layers of the vascular wall. Upper limb imaging showed preserved radial flow. This case illustrates coronary embolization secondary to radial arterial wall injury during transradial PCI. Difficult catheter advancement during transradial PCI may cause radial arterial wall avulsion with subsequent coronary embolization. Unexpected angiographic filling defects should prompt consideration of nonthrombotic embolic mechanisms and rapid bailout strategies.
Left main (LM) ST-elevation myocardial infarction (STEMI) is typically managed with urgent stent implantation according to current guidelines. Yet intravascular imaging has shown that plaque erosion may permit alternative strategies. A 43-year-old man with anterior STEMI and hypotension underwent angiography, revealing distal LM thrombus extending into the left anterior descending/left circumflex arteries. Postaspiration optical coherence tomography (OCT) confirmed plaque erosion without rupture. Following lesion preparation, a drug-coated balloon-only strategy was applied to all 3 vessels. Follow-ups at 20 days and 15 months confirmed vessel patency, satisfactory healing, and no recoil or neoatherosclerosis. Current guidelines lack support for stent-free LM STEMI intervention. However, OCT-guided management allows for individualized treatment in young patients by characterizing lesion biology. Long-term follow-up demonstrates successful vessel healing, suggesting that avoiding permanent metallic scaffolds in selected cases may mitigate lifelong stenting risks. OCT can identify plaque erosion and guide stent-free percutaneous coronary intervention in select left main STEMI cases, offering durable healing and flexibility in antithrombotic therapy.
Infective endocarditis following transcatheter aortic valve replacement and mitral valve transcatheter edge-to-edge repair poses a significant clinical challenge, necessitating a multidisciplinary approach for timely diagnosis and appropriate management that may include surgical intervention for patients of less than prohibitive risk. Although advances in imaging modalities, including 18F-fluorodeoxyglucose positron emission tomography/computed tomography, show promise in enhancing diagnostic accuracy, standardized guidelines for management are lacking, especially regarding the appropriateness of antimicrobial suppression therapy. Future research should focus on refining diagnostic criteria, optimizing therapeutic strategies, and improving patient outcomes in these populations.
Although recent large-scale randomized trials have demonstrated that sodium-glucose cotransporter 2 (SGLT2) inhibitors following acute myocardial infarction (AMI) are safe, the mechanisms underlying their potential cardioprotective effects remain poorly understood. The aim of this study was to evaluate the effects of SGLT2 inhibitor therapy on myocardial injury and left ventricular (LV) remodeling in AMI patients undergoing percutaneous coronary intervention (PCI), using cardiac magnetic resonance imaging (CMR). In this prospective, open-label, randomized controlled trial, patients ≥18 years of age at high risk for heart failure after successful PCI for AMI were randomly assigned to receive empagliflozin 10 mg once daily or not. The primary endpoint was infarct size (% LV mass) assessed by CMR at 6-month follow-up. The coprimary endpoint was a difference in LV end-systolic volume measured by CMR between baseline and 6 months (ΔLV end-systolic volume). A total of 200 patients underwent randomization, with 100 assigned to the SGLT2 inhibitor group and 100 assigned to the control group. CMR assessments for 6 months were available for 169 patients (84.0%) of the total study population (89 patients in the SGLT2 inhibitor group and 80 patients in the control group). Compared with control, SGLT2 inhibition did not reduce infarct size (% LV mass) at 6-month follow-up (SGLT2 inhibitor vs control, 12.5% [8.5%-20.9%] vs 12.9% [5.0%-20.7%]; P = 0.92). There was no significant difference in ΔLV end-systolic volume between the 2 groups (SGLT2 inhibitor vs control, -3.3% [-19.7% to 13.6%] vs -6.8% [-23.4% to 10.6%]; P = 0.40). Among patients with AMI undergoing PCI who were at high risk for heart failure, SGLT2 inhibitor use did not reduce infarct size or facilitate reverse LV remodeling at 6 months as assessed by CMR. (Peri-Treatment of SGLT-2 Inhibitor on Myocardial Infarct Size and Remodeling Index in Patients With Acute Myocardial Infarction and High Risk of Heart Failure Undergoing Percutaneous Coronary Intervention [PRESTIGE-AMI]; NCT04899479).
Isolated tricuspid regurgitation (TR) is an independent mortality predictor. Despite improved surgical outcomes, complex tricuspid valve anatomy makes precise repair challenging. We evaluated the utility of computed tomography (CT) with radial-sectional view (RSV) for procedural planning and postoperative assessment. Four patients with severe TR underwent surgical repair. Preoperative CT-RSV, generated using standard imaging workstations, provided a comprehensive anatomical map. Radial planes centered on the septal leaflet visualized coaptation gaps with anterior/posterior leaflets, the location of prolapse, and the severity of tethering. This "virtual surgery" enabled individualized strategies. In all cases, tricuspid valve repair was performed according to CT-based planning, resulting in marked reduction of TR. CT-RSV provides a reproducible and accessible roadmap. By detailing septal-centric anatomy, it facilitates clearer visualization of leaflet and subvalvular structures compared with transesophageal echocardiography and enables reproducible quantitative assessment, thereby supporting individualized surgical planning and informing treatment strategy selection.
In patients with peripheral artery disease and chronic limb-threatening ischemia (CLTI), there is a lack of effective treatment options when arterial revascularization is not feasible. In preclinical models, ultrasound (US) cavitation of microbubble contrast agents can augment limb perfusion through multiple shear-mediated vasoactive pathways. This study aims to evaluate whether US cavitation increases limb tissue perfusion and accelerates healing of ischemic ulcers in patients with CLTI. Patients (n = 24) with CLTI and chronic nonhealing ulcers were randomized to standard care or cavitation therapy, which was performed every 3 days × 6. Definity was infused, and US (1.3 MHz, mechanical index: 1.3) was performed for 15 minutes at the wound site and simultaneously over the ipsilateral femoral artery to leverage conducted vasodilation. The primary endpoints at day 18 were change in ulcer area and ulcer microvascular perfusion. Complete or partial ulcer healing occurred in almost all cavitation-treated patients, with a significant reduction in ulcer size only in the cavitation group and not the control group (median: -36.6% [Q1-Q3: -15.4% to -65.0%] vs +3.4% [Q1-Q3: -14.7% to +20.0%]; P = 0.039). In cavitation-treated patients, there was a 2.5-fold increase in ulcer margin perfusion, a 1.6-fold increase in functional microvascular density on super-resolution US microangiography, and a 1.8-fold increase in femoral artery flow (P < 0.05 for all parameters). These flow parameters did not increase in control patients. In patients with CLTI and nonhealing ulcers, periodic therapy with cavitation of microbubbles using US parameters approved for diagnostic imaging promotes healing of ischemic ulcers, likely through improvement in tissue perfusion and microvascular vascularity.
Left ventricular systolic dysfunction (LVSD) is the main predictor of mortality in Chagas disease (ChD). Although LVSD can be treated with affordable medications, its diagnosis relies on cardiac imaging, which is often unavailable in resource-limited settings. The objective of the study was to evaluate an artificial intelligence-enabled electrocardiogram (AI-ECG) for detecting LVSD and predicting mortality and incident LVSD in ChD. A previously developed AI-ECG LVSD model was fine-tuned in an external ChD sample and applied to SaMi-Trop, a Brazilian prospective ChD cohort. Diagnostic performance for LVSD, confirmed by echocardiography, was compared with N-terminal pro-B-type natriuretic peptide (NT-proBNP). Prognostic performance for all-cause mortality at 2 and 9 years was assessed using Cox models, and incident LVSD over 7 years using log-binomial models. Among 1,304 participants, AI-ECG showed high accuracy for LVSD detection (area under the receiver operating characteristic curve: 0.89; 95% CI: 0.85-0.93), similar to NT-proBNP (area under the receiver operating characteristic curve: 0.90; 95% CI: 0.87-0.93; P = 0.52). Among 1,547 patients with Chagas cardiomyopathy, AI-ECG predicted all-cause mortality at 2 and 9 years with performance comparable to NT-proBNP (9-year C-index: 0.78; 95% CI: 0.74-0.82 vs 0.77; 95% CI: 0.74-0.79). AI-ECG could replace NT-proBNP in an established ChD mortality risk score with minor loss of accuracy. Incident LVSD occurred in 8.4% over 7.3 years, and AI-ECG predicted incident LVSD with performance similar to NT-proBNP. A fine-tuned AI-ECG model showed high accuracy for LVSD detection and was independently associated with long-term mortality and incident LVSD. Despite slightly lower performance than NT-proBNP, it may serve as a substitute in settings where NT-proBNP is unavailable. (Longitudinal Study of Patients With Chronic Chagas Cardiomyopathy in Brazil [SaMi_Trop Project] [SaMi-Trop]; NCT02646943).