A simplified clinical electrocardiogram score for the prediction of cardiovascular mortality

Abstract

Background: Electrocardiogram (ECG) scores have been demonstrated to predict CV mortality but they are rarely utilized clinically. Objective: Develop a simple score consisting of adding classical ECG abnormalities to make the ECG a more convenient prognostic tool. Methods: Resting ECGs of 29,320 outpatient male veterans from the Palo Alto Veteran Affairs Healthcare System (PAVHS) collected between 1987 and 2000 were computer analyzed with an average follow-up of 7.5 y. Twelve classic ECG abnormalities were chosen on the basis of prevalence and corresponding relative risks, including left and right bundle branch block, diagnostic Q waves, intraventricular conduction defect, atrial fibrillation, left atrial abnormality, left and right axis deviation, left and right ventricular hypertrophy, ST depression, and abnormal QTc interval. A simple score derived from the summation of these criteria was then entered into an age and heart rate adjusted Cox analysis. Results: There was a progressive increase in risk of death as the number of ECG abnormalities increased. The relative risks for 1, 2, 3, 4, and 5 ECG abnormalities were 1.8 (CI 1.6-2.0), 2.4 (CI 2.2-2.7), 3.6 (CI 3.2-4.1), 4.5 (CI 3.8-5.4), and 6.0 (CI 4.7-7.8) respectively (p<0.001). The age-adjusted hazard ratio for CV mortality was 6.0 when there were five or more ECG abnormalities present. Conclusion: Summing the number of classical ECG abnormalities provides a powerful predictor of CV mortality independent of age, standard risk factors, and clinical status. © 2009 Wiley Periodicals, Inc.