The emergency department ECG and immediately life-threatening complications in initially uncomplicated suspected myocardial ischemia

The emergency physician's disposition of patients with suspected myocardial ischemia is currently debated; some physicians believe that a subgroup of patients can be managed safely outside the coronary care unit. Clinical predictors are needed in assessing the patient with suspected myocardial ischemia to help identify this subgroup. Through a retrospective cohort study, we investigated the value of the initial emergency department ECG in discriminating between chest pain patients with low and high risk for immediately life-threatening complications. Two hundred eleven initially uncomplicated consecutive coronary care unit admissions with suspected unstable angina or myocardial infarction were studied. Patient outcome, including the incidence of myocardial infarction, complications, and mechanical and pharmacologic interventions, was reviewed. Immediately life-threatening complications included ventricular fibrillation, ventricular tachycardia, shock, 2° and 3° block, and death. Mechanical interventions included electrocardioversion or defibrillation, endotracheal intubation, intra-aortic ballon pump, Swan-Ganz catheter, or pacemaker insertion. Pressors, antiarrhythmics, and vasodilators were the reviewed pharmacologic interventions. A positive ECG was defined by the presence of ST elevation or depression, T wave inversion, left ventricular hypertrophy, left bundle branch block, paced rhythm, or new Q waves. All other ECG inter-pretations were considered negative. Patients were divided into two groups based on this initial emergency physician ECG interpretation and their complication incidences compared. Of the 211 patients, 96 had a positive ECG; 115 had negative ECGs. Patients with positive ECGs were older, had a greater history and concurrent incidence of myocardial infarction, and more complications and intensive interventions. No patients in the negative ECG group had immediately life-threatening complications, as compared with an incidence of 14.6% in the positive ECG group ( P < .001). The incidence of mechanical interventions was 4.3% in the negative group, compared with 20.8% in the positive group ( P < .001). The incidence of pharmacological interventions was 12.2% in the negative group, compared with 34.4% ( P < .001) in the positive ECG group. These data suggest that the initial emergency department ECG can be used to identify a low-risk group for life-threatening complications and the need for interventions requiring the coronary care unit. Pa