Artificial Intelligence–Enabled Electrocardiographic Detection of Severe Aortic Stenosis Leading to Transcatheter Aortic Valve Replacement

BACKGROUND: EchoNext is an artificial intelligence (artificial intelligence)-enabled electrocardiographic (ECG) model validated to detect unrecognized structural heart disease. FIRST-IN-HUMAN/EARLY REPORTS SUMMARY: An 84-year-old woman presented after a fall and was found to have a left femur fracture. An ECG was ordered which was analyzed by EchoNext. Despite the absence of cardiac symptoms, the ECG was flagged to suggest structural heart disease, prompting cardiology consultation and evaluation. Transthoracic echocardiogram revealed a reduced left ventricular ejection fraction and severe aortic stenosis (AS). After orthopedic surgery with cardiac anesthesia, the patient underwent transcatheter aortic valve replacement. DISCUSSION/NOVELTY: Early identification of severe AS altered perioperative management in an otherwise asymptomatic patient. To our knowledge, this is the first report of an artificial intelligence-enabled ECG algorithm detecting undiagnosed severe AS that led to transcatheter aortic valve replacement. TAKE-HOME MESSAGE: AI-enabled ECG models may enable earlier detection of clinically silent structural heart disease.