ECG-Derived Synthetic Left Ventricular Mechanics Predict Mortality After Acute Myocardial Infarction and Fatal Myocardial Infarction in the General Population
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Left ventricular (LV) mechanics predict outcomes after acute myocardial infarction (AMI). Recovering them from the 12-lead electrocardiogram (ECG) could enable risk assessment wherever an ECG is acquired.
Objectives
To test whether ECG-derived synthetic LV mechanics predict mortality after AMI and myocardial infarction (MI)-specific mortality in the general population beyond the GRACE score and ejection fraction (EF).
Methods
Synthetic peak systolic (s′) and early diastolic (e′) mitral annular velocities and mitral annular plane systolic excursion (MAPSE) were derived from the ECG by a validated generative adversarial network, applied without retraining and blinded to outcome. We studied an AMI cohort (n=1,065; Cox regression) and a population cohort (n=219,564; Fine-Gray models), testing incremental value over GRACE, three artificial intelligence (AI)-ECG reduced-EF models, and measured EF.
Results
In the AMI cohort (109 deaths; median follow-up 1.44 years), all three parameters predicted all-cause mortality, but only synthetic e′ remained independent of clinical covariates (HR=0.90 per 1 cm/s; 95% CI 0.84-0.98) or GRACE (HR=0.90; 0.83-0.97). Among 1,017 patients discharged alive (61 deaths), e′ improved discrimination beyond GRACE (C-index 0.67 to 0.70) and remained independent of AI-ECG-predicted and measured EF (HR 0.84-0.86 per 1 cm/s; all P≤0.004). In the population cohort (817 MI deaths; median follow-up 3.42 years), all three predicted MI-specific mortality after adjustment for age, sex, and ECG abnormalities (each P<0.001), most strongly MAPSE (sHR=0.49 per 1 cm; 95% CI 0.37-0.65).
Conclusions
ECG-derived synthetic LV mechanics predict mortality after AMI and fatal MI in the general population, adding information beyond GRACE and EF.
