Dilated cardiomyopathy (DCM) is a common and serious heart disease characterized by left ventricular enlargement and impaired pumping function, with adverse prognosis (including heart failure, arrhythmia, heart-related hospitalization, and death) being a major concern for patients. Currently, a critical gap exists in accurately predicting which DCM patients are at high risk of these severe outcomes, limiting targeted clinical care.

This observational, non-invasive study aims to develop and validate a clinical prediction model for early risk warning of adverse prognosis in DCM patients. The model integrates multi-parameter stress perfusion cardiac magnetic resonance (MP stress perfusion CMR)-a safe, high-resolution imaging technique that assesses cardiac structure, function, blood perfusion, and tissue damage under mild stress-and standard clinical data (e.g., age, gender, blood pressure, and routine heart test results).

The model will be trained and tested using follow-up data from hundreds of DCM patients, with the analysis identifying patterns in CMR and clinical data associated with adverse outcomes. Once validated for accuracy, the model will provide doctors with personalized risk scores to prioritize care for high-risk patients (e.g., early intervention, close monitoring) and avoid over-treatment for lower-risk individuals.

Beyond clinical application, the study will enhance understanding of DCM progression, laying the groundwork for improved diagnostic tools, more effective treatments, and better strategies to prevent DCM-related complications, ultimately improving patient quality of life and reducing mortality.

The study focuses on a common but serious heart condition called dilated cardiomyopathy (DCM), which affects millions of people worldwide. Dilated cardiomyopathy is a disease where the heart's main pumping chamber (the left ventricle) becomes enlarged and weakened, making pumping function harder for the heart to pump blood efficiently to the rest of the body. For patients with DCM, the biggest concern is the risk of "adverse prognosis"-a category includes serious outcomes like heart failure, abnormal heart rhythms, hospitalizations related to heart problems, or even death. Currently, doctors face challenges in accurately predicting which DCM patients are more likely to experience these severe outcomes, a limitation restricting the ability to provide timely, targeted care to those at highest risk.

To address the gap, the investigators aim to develop and test a "clinical prediction model"-a tool that combines medical data to predict the likelihood of adverse outcomes in DCM patients. The key innovation of the model is the use of a powerful medical imaging technique called multi-parameter stress perfusion cardiac magnetic resonance (MP stress perfusion CMR), combined with standard clinical information about patients.

First, the terms are explained in simple language. Cardiac magnetic resonance (CMR) is a safe, non-invasive imaging test that creates detailed pictures of the heart's structure and function. "Stress perfusion" means that during the CMR scan, the participant's heart is placed under a mild form of stress (usually by administration of a medication that increases heart rate, similar to light exercise) to observe blood flow through the heart muscle during increased workload. "Multi-parameter" refers to the collection of multiple types of information from the CMR scan, such as myocardial contractility, myocardial perfusion, and any damage or scarring in cardiac tissue.

The goal of the study is to use detailed CMR data, along with other basic clinical information (e.g., patient age, gender, blood pressure, and results from standard heart tests), to construct a prediction model. The model will be trained using data from hundreds of DCM patients with existing follow-up records. The investigators will analyze the records to determine which patients experienced adverse outcomes and to identify patterns in CMR and clinical data associated with those outcomes.

Once the model is constructed and tested for accuracy, the tool can be utilized by doctors in clinical practice to assist in the assessment of DCM patients. For example, when a patient is diagnosed with DCM, doctors can perform the CMR scan, input the data into the model, and obtain a personalized risk score indicating the probability of future severe cardiac events in that patient. The resulting information will support clinical decision-making: the scores will allow doctors to prioritize care for high-risk patients-such as early treatment initiation, more frequent monitoring, or medication adjustment-to prevent adverse outcomes. For lower-risk patients, the model can help avoid unnecessary, costly tests or over-treatment.

Beyond individual patient benefits, the study also has broader value for public health and medical research. By better defining the factors (from CMR and clinical data) most closely linked to poor outcomes in DCM, new insights can be gained regarding disease progression. Such insights may lead to improved diagnostic tools, more effective treatments, and enhanced strategies for preventing DCM-related complications in the long term.

Of particular note, the study is observational and non-invasive. All patients receive standard medical care for DCM, and the CMR scan is a routine test already applied in clinical practice. The study will follow patients over time to validate the prediction model, ensuring reliable performance across different subgroups of DCM patients (e.g., varying age, gender, or disease severity).

In summary, the study seeks to bridge a critical gap in the care of DCM patients by creating a reliable, easy-to-use prediction tool based on advanced CMR imaging. By early identification of patients at high risk of adverse outcomes, quality of life can be improved, hospitalizations reduced, and lives potentially saved-while also advancing understanding of the complex heart disease for the benefit of future patients.

Inclusion Criteria

1.An elevated left ventricular end-diastolic volume indexed to body surface area and reduced LVEF, compared with published age- and gender-specific reference values Exclusion Criteria

1. significant coronary artery disease (CAD), defined as a stenosis of ˃50% in a major coronary artery
2. infiltrative disease
3. valvular cardiomyopathy
4. arrhythmogenic cardiomyopathy
5. congenital heart disease

During the study period, patients with dilated cardiomyopathy (DCM) who were seen in the cardiology departments of Shandong Provincial Hospital, Jinan Central Hospital, and Beijing Anzhen Hospital, or referred for cardiac magnetic resonance (CMR) assessment, were prospectively enrolled in the registry at the time of scanning.

Individual participant data (IPD) will not be made available to external researchers.

The dataset contains sensitive clinical, imaging, and longitudinal prognostic information from patients with dilated cardiomyopathy.

Broad sharing of IPD may compromise patient privacy and confidentiality, violate informed consent restrictions, and increase the risk of re-identification.

In addition, the multiparametric CMR and artificial intelligence models rely on integrated institutional data that have not been de-identified to a level suitable for unrestricted public or third-party sharing.

Therefore, IPD will be retained securely within the study group and will not be shared externally.