Electrocardiographic ST-segment elevation: The diagnosis of acute myocardial infarction by morphologic analysis of the ST segment

Acute myocardial infarction (AMI) is one of many causes of ST-segment elevation (STE) in emergency department (ED) chest pain (CP) patients. The morphology of STE may assist in the correct determination of its cause, with concave patterns in non-AMI syndromes and non-concave waveforms in AMI. Objectives: To determine the impact of STE morphologic analysis on AMI diagnosis and the ability of this technique to separate AMI from non-infarction causes of STE. Methods: The electrocardiograms (ECGs) of consecutive ED adult CP patients (with three serial troponin I determinations) were interpreted in two-step fashion by six attending emergency physicians (EPs): 1) the determination of STE by three EPs followed by 2) STE morphologic analysis (either concave or non-concave) in those patients with STE. The impact of STE morphology analysis was investigated in the identification of AMI and non-AMI causes of STE. Acute myocardial infarction was diagnosed by abnormal serum troponin I values (>0.1 mg/dL) followed by a rise and fall of the serum marker; STE diagnoses of non-AMI causes were determined by medical record review. Interobserver reliability concerning STE morphology was determined. Study inclusion criteria included at least three troponin values performed in serial fashion no more frequently than every three hours, initial ED ECG, ED diagnosis, and final hospital diagnosis. Results: Five hundred ninety-nine CP patients were entered in the study, with 171 (29%) individuals having STE on their ECGs. Of the 171 patients who had STE, 56 had AMI, 50 had unstable angina pectoris (USAP), and 65 had non-coronary final diagnoses. Forty-nine patients had non-concave STE, 46 with AMI and three with USAP; no patient with a non-coronary diagnosis had a non-concave STE morphology. The sensitivity and specificity of the non-concave STE morphology for AMI diagnoses were 77% and 97%, respectively; the positive and negative predictive values for non-concave morphology in AMI diagnoses were 94% and 88%, respectively. Interobserver reliability in the STE morphology determination revealed a kappa coefficient of 0.87. Conclusions: A non-concave STE morphology is frequently encountered in AMI patients. While the sensitivity of this pattern for AMI diagnosis is not particularly helpful, the presence of this finding in adult ED chest pain patients with STE strongly suggests AMI. This technique produces consistent results among these EPs.