ECG-Based Algorithm Predicts Long-Term Survival After Cardiac Surgery

WEDNESDAY, Dec. 22, 2021 (HealthDay News) — An electrocardiography-based artificial intelligence (AI) algorithm that predicts severe ventricular dysfunction can predict long-term mortality among patients undergoing valve and/or coronary bypass surgery, according to a study published in the December issue of Mayo Clinic Proceedings.

Abdulah A. Mahayni, from the Mayo Clinic in Rochester, Minnesota, and colleagues assessed whether an electrocardiography-based AI algorithm developed to detect severe ventricular dysfunction (left ventricular ejection fraction [LVEF] ≤35 percent) independently predicts long-term mortality after cardiac surgery among patients without severe ventricular dysfunction (LVEF >35 percent). The analysis included 20,627 patients who underwent valve or coronary bypass surgery (1993 to 2019) and had documented LVEF >35 percent.

The researchers found that 83.0 percent had a normal AI-electrocardiogram (ECG) screen, and 17.0 percent had an abnormal AI-ECG screen. In patients with a normal AI-ECG screen, probability of survival at five and 10 years was 86.2 and 68.2 percent, respectively, versus 71.4 and 45.1 percent, respectively, in those with an abnormal screen. An abnormal AI-ECG screen was independently associated with a higher all-cause mortality overall (hazard ratio [HR], 1.31) and in subgroups of isolated valve surgery (HR, 1.30), isolated coronary artery bypass grafting (HR, 1.29), and combined coronary artery bypass grafting and valve surgery (HR, 1.19).

“Our study finds there is a clear correlation between long-term mortality and a positive AI ECG screen for reduced ejection fraction among patients without apparent severe cardiomyopathy,” a coauthor said in a statement.

Abstract/Full Text