Using Electrocardiogram to Assess Diastolic Function and Prognosis in Mitral Regurgitation.

Area of Science:

• Cardiology
• Artificial Intelligence in Medicine
• Medical Diagnostics

Background:

• Assessing left ventricular diastolic function (LVDF) in patients with significant mitral regurgitation (MR) is clinically challenging.
• An artificial intelligence-enabled electrocardiogram (AI-ECG) algorithm has been developed and validated for estimating LVDF.

Purpose of the Study:

• To evaluate the association between AI-ECG-derived LVDF grades and the risk of all-cause mortality in patients with significant MR.
• To assess the prognostic value of AI-ECG LVDF grading for myocardial disease (MD) in the context of MR.

Main Methods:

• Retrospective analysis of patients with significant MR and electrocardiogram (ECG) within 14 days, from Mayo Clinic (2001-2023).
• Patients categorized into three myocardial disease (MD) grades (MD-1, MD-2, MD-3) based on AI-ECG LVDF.
• Multivariable survival analysis was performed to assess the association between MD grades and all-cause mortality.

Main Results:

• A total of 4,019 patients with significant MR were included; 40.7% died during a median follow-up of 3.5 years.
• Higher AI-ECG LVDF grades were independently associated with increased all-cause mortality risk (MD-2: aHR 1.99; MD-3: aHR 2.65).
• These associations remained consistent across sensitivity analyses, including adjustments for mitral valve intervention.

Conclusions:

• In patients with significant mitral regurgitation, LVDF grading using AI-ECG is an independent predictor of all-cause mortality.
• AI-ECG LVDF assessment offers valuable prognostic information for risk stratification in this patient population.