HealthDay News — For patients presenting to the emergency department with dyspnea, an artificial intelligence-enabled electrocardiogram (AI-ECG) can identify left ventricular systolic dysfunction (LVSD; defined as left ventricular ejection fraction ≤35 percent) with high accuracy, according to a study published online Aug. 4 in Circulation: Arrhythmia and Electrophysiology.
Demilade Adedinsewo, M.D., M.P.H., from the Mayo Clinic in Jacksonville, Florida, and colleagues retrospectively applied a validated AI-ECG algorithm to a cohort of 1,606 patients aged ≥18 years. Participants had at least one standard 12-lead ECG acquired on the date of the emergency department visit and an echocardiogram performed within 30 days.
The researchers found that LVSD was identified by AI-ECG with an area under the receiver operating characteristic curve (AUC) of 0.89 and accuracy of 85.9 percent. The sensitivity and specificity were 74 and 87 percent, while the negative and positive predictive values were 97 and 40 percent, respectively. The AUC and accuracy for identifying an ejection fraction of <50 percent were 0.85 and 86 percent, while sensitivity and specificity were 63 and 91 percent, respectively. LSVD was identified by N-terminal proB-type natriuretic peptide at a cutoff of >800 with an AUC of 0.80.
“The application of an AI-ECG algorithm in the emergency department could improve diagnostic accuracy, facilitate appropriate disposition and provide an avenue to identify high risk patients early and link them to appropriate cardiovascular care,” the authors write.