Prediction of atrial and ventricular fibrillation complicating myocardial infarction from admission data: a...

This study set out to examine prospectively two logistic formulae based on admission clinical data to predict ventricular or atrial fibrillation complicating acute myocardial infarction. A prospective study of 87 consecutive patients with acute transmural myocardial infarction was conducted. The formula for predicting ventricular fibrillation from the diastolic blood pressure, degree of ST-segment elevation, and QTc had a sensitivity of 93%, specificity of 83%, and a predictive value for an abnormal test of 62% (13 of 14 patients who developed ventricular fibrillation were identified). The formula for predicting atrial fibrillation from the age of the patient, a history of heart failure, systolic blood pressure, and four electrocardiographic parameters had a sensitivity of 78%, specificity of 85%, and a predictive value of 67% (14 of 18 patients identified). Our study shows that patients with myocardial infarction who are liable to develop ventricular or atrial fibrillation can be identified on admission from simple clinical data.