SCAN-MRI: Cardiac MRI-integrated Risk Score for Predicting Cardiovascular Events in Type 2 Diabetes Mellitus

Area of Science:

• Cardiology
• Radiology
• Medical Imaging
• Diabetes Mellitus Research

Background:

• Patients with diabetes mellitus (DM) face elevated risks of cardiovascular complications.
• Current risk assessment tools for DM primarily use clinical factors, neglecting cardiac structure and function insights from imaging biomarkers.

Purpose of the Study:

• To develop and validate a predictive model for cardiovascular outcomes in type 2 DM patients using cardiac MRI.
• To compare the performance of the novel MRI-based model against established clinical risk scores.

Main Methods:

• Prospective and retrospective enrollment of DM participants who underwent cardiac MRI.
• Development of a multivariable Cox regression model incorporating clinical and cardiac MRI parameters.
• Validation of the SCAN-MRI risk score against WATCH-DM and TIMI-HF-DM models using AUC and C-index.

Main Results:

• The SCAN-MRI model demonstrated superior predictive performance (AUC, 0.76) compared to WATCH-DM (AUC, 0.66) and TIMI-HF-DM (AUC, 0.65).
• Integration of cardiac MRI markers significantly enhanced the discriminative ability of existing risk models.
• The model showed good predictive performance in an external test set (C-index, 0.71).

Conclusions:

• A cardiac MRI-based risk model integrating clinical and imaging data offers improved prediction of adverse cardiovascular outcomes in DM patients.
• The SCAN-MRI score provides a more accurate and comprehensive risk assessment compared to current clinical risk models.