Diagnostic Accuracy of a New Cardiac Electrical Biomarker for Detection of Electrocardiogram Changes Suggestive of Acute Myocardial Ischemic Injury

• Published in: Annals of noninvasive electrocardiology Vol. 19; no. 2; pp. 129 - 144
• Main Authors: Schreck, David M., Fishberg, Robert D.
• Format: Journal Article
• Language: English
• Published: United States Blackwell Publishing Ltd 01.03.2014; John Wiley & Sons, Inc; John Wiley and Sons Inc
• Subjects: Accuracy; acute myocardial infarction; acute myocardial ischemia; biomarker; Biomarkers; Case-Control Studies; chest pain; Confidence intervals; Electrocardiography; Electrocardiography - methods; Electrocardiography - statistics & numerical data; Female; Humans; Hypothesis testing; Male; Middle Aged; Myocardial Infarction - diagnosis; Myocardial Infarction - physiopathology; myocardial injury; Original; Reproducibility of Results; Retrospective Studies; ROC Curve; Sensitivity and Specificity; Signal Processing, Computer-Assisted; Single-Blind Method; acute myocardial ischemia; myocardial injury; biomarker; chest pain; acute myocardial infarction; electrocardiography
• ISSN: 1082-720X; 1542-474X; 1542-474X
• DOI: 10.1111/anec.12109

Objective A new cardiac “electrical” biomarker (CEB) for detection of 12‐lead electrocardiogram (ECG) changes indicative of acute myocardial ischemic injury has been identified. Objective was to test CEB diagnostic accuracy. Methods This is a blinded, observational retrospective case‐control, noninferiority study. A total of 508 ECGs obtained from archived digital databases were interpreted by cardiologist and emergency physician (EP) blinded reference standards for presence of acute myocardial ischemic injury. CEB was constructed from three ECG cardiac monitoring leads using nonlinear modeling. Comparative active controls included ST voltage changes (J‐point, ST area under curve) and a computerized ECG interpretive algorithm (ECGI). Training set of 141 ECGs identified CEB cutoffs by receiver‐operating‐characteristic (ROC) analysis. Test set of 367 ECGs was analyzed for validation. Poor‐quality ECGs were excluded. Sensitivity, specificity, and negative and positive predictive values were calculated with 95% confidence intervals. Adjudication was performed by consensus. Results CEB demonstrated noninferiority to all active controls by hypothesis testing. CEB adjudication demonstrated 85.3–94.4% sensitivity, 92.5–93.0% specificity, 93.8–98.6% negative predictive value, and 74.6–83.5% positive predictive value. CEB was superior against all active controls in EP analysis, and against ST area under curve and ECGI by cardiologist. Conclusion CEB detects acute myocardial ischemic injury with high diagnostic accuracy. CEB is instantly constructed from three ECG leads on the cardiac monitor and displayed instantly allowing immediate cost‐effective identification of patients with acute ischemic injury during cardiac rhythm monitoring.