Artificial intelligence-enhanced electrocardiogram diastolic function grade predicts post-septal myectomy mortality in hypertrophic cardiomyopathy.

Abstract

Diastolic dysfunction is an important pathophysiologic feature of hypertrophic cardiomyopathy that is often challenging to determine noninvasively. This study investigated whether a novel artificial intelligence-enabled electrocardiography (AI-ECG)-determined diastolic function grade is associated with early postoperative outcomes and long-term mortality in patients with obstructive hypertrophic cardiomyopathy (oHCM) following septal myectomy. We analyzed 1953 patients undergoing transaortic septal myectomy and stratified their outcomes by grade of diastolic function as determined by a validated deep-learning AI algorithm applied on each patient's preoperative 12-lead electrocardiogram. Outcomes of interest included early postoperative morbidity, hospital length of stay, and all-cause mortality. Patients with AI-ECG diastolic function grade 3 had a higher prevalence of comorbidities, including dyslipidemia, hypertension, congestive heart failure, and atrial fibrillation/flutter (AF). AI-ECG diastolic function grade significantly correlated with echocardiographic measures, including septal and posterior wall thickness, left atrial volume index, left ventricular mass index, maximum left ventricular outflow tract gradient, medial E/e', medial e', mitral and tricuspid regurgitation. Patients with AI-ECG diastolic (dys)function grades 2 and 3 had an increased risk of late mortality (hazard ratio [HR], 1.73; 95% confidence interval [CI], 1.05-2.87 and HR, 2.45; 95% CI, 1.29-4.67, respectively). Patients with diastolic function grade 3 were more likely to have postoperative AF (odds ratio, 1.63; 95% CI, 1.04-2.56) and had longer hospital length of stay (+0.87 days; P < .01). Higher AI-ECG diastolic grade is associated with an increased risk of postoperative AF, longer hospital length of stay, and worse long-term survival. AI-ECG diastolic grade may be a useful tool for risk stratification of patients with oHCM undergoing septal myectomy.