Explainable advanced electrocardiography at rest for ruling out myocardial ischemia on stress echocardiography

Abstract

Background

Stress echocardiography (SE) is relatively resource intensive and has a low incidence of abnormal tests for detecting coronary artery disease (CAD) in low-to-intermediate risk patients. This study aimed to derive and determine the diagnostic performance of a resting advanced electrocardiography (A-ECG) score for detecting inducible myocardial ischemia on SE in patients with low-to-intermediate risk stable chest pain.

Methods

Patients were included if they presented with low-to-intermediate risk stable chest pain to the emergency department, had acute coronary syndrome ruled out by electrocardiography (ECG) and high-sensitivity troponin, and subsequently underwent outpatient SE. Patients were excluded if they had known CAD or confounders on resting ECG. A-ECG was retrospectively applied to a standard resting 12‑lead ECG and a multivariable logistic regression score was derived to predict myocardial ischemia on SE.

Results

Among 292 patients (51 % male, age 58 ± 14 years), 24 (8 %) exhibited inducible myocardial ischemia on SE. A 3-parameter A-ECG score had an area under the receiver-operating characteristic curve (AUC [bootstrapped 95 % confidence interval]) of 0.85 [0.75–0.93], sensitivity 92 [67–100]%, specificity 67 [64–94]%, positive predictive value 22 [20–55]%, negative predictive value 99 [96–100]%, positive likelihood ratio 2.8 [2.5–12.0] and inverse negative likelihood ratio 8.1 [2.5–18.0].

Conclusions

An A-ECG score had a good overall diagnostic performance and excellent performance for ruling out inducible myocardial ischemia on SE. This supports the use of an A-ECG score to triage and improve the selection of patients with low-intermediate risk stable chest pain that should undergo further testing with SE.