Early detection of acute myocardial infarction in patients presenting with chest pain and nondiagnostic ECGs: Serial CK-MB sampling in the emergency department

Study objectives: Patients presenting to the emergency department with chest discomfort are a difficult problem for emergency physicians. Nearly 50% of patients with acute myocardial infarction (AMI) will initially have nondiagnostic ECGs on ED presentation. The purpose of this study was to determine if patients with AMI having nondiagnostic ECGs could be identified using new immunochemical assays for serial CK-MB sampling in the ED. Design: Chest pain patients, more than 30 years old, with pain not caused by trauma or explained by radiographic findings, were eligible for the study. Serial serum samples were drawn on ED presentation (zero hours) and three hours after presentation, then analyzed for CK-MB using four immunochemical methods and electrophoresis. Standard World Health Organization criteria were used to establish the diagnosis of AMI, including new Q-wave formation or elevation of standard in-hospital serum cardiac enzyme markers. Setting: A tertiary cardiac care community hospital. Measurements and main results: The serum from 183 patients hospitalized for possible ischemic chest pain was collected and analyzed. Thirty-one of 183 patients (17%) were found to have AMI by standard in-hospital criteria. Sixteen of the 31 patients (52%) with AMI had nondiagnostic ECGs on presentation. Immunochemical determination of serial CK-MB levels provided a sensitive and specific method for detecting AMI in patients within three hours after ED presentation compared with standard electrophoresis. The four immunochemical methods demonstrated a range in sensitivity from 50% to 62.1% on ED presentation versus 92% to 96.7% three hours later. The immunochemical tests demonstrated specificities ranging from 83.0% to 96.4% at three hours, with three of the four tests having specificities of 92% or greater. Electrophoresis had a sensitivity of 34.5% on ED presentation, increasing to 76.9% at three hours, with a specificity of 98.6%. Conclusions: Immunochemical CK-MB methods allowed rapid, sensitive detection of AMI in the ED. Early detection of AMI offers many potential advantages to the emergency physician. Early detection of AMI, while the patient is in the ED, could direct disposition of this potentially unstable patient to an intensive care setting. Such information may prevent the ED discharge of patients with AMI having nondiagnostic ECGs. The diagnosis of AMI within a six-hour period after symptom onset may allow thrombolytic therapy to be given to patients with AMI not having diagnostic ECGs. This study served as a pilot trial for a multicenter study of the Emergency Medicine Cardiac Research Group, which is currently ongoing.