Computer-assisted versus unassisted analysis of the exercise electrocardiogram in patients without myocardial...

Computer-assisted interpretation of the exercise electrocardiogram has been advocated to improve the accuracy of diagnosing coronary artery disease. Its accuracy was compared with a blinded visual interpretation of exercise-induced ST depression in 271 consecutive subjects without prior myocardial infarction who were referred for coronary angiography. The sensitivity of the visual and computer readings was 0.51 and 0.51, respectively, at a specificity of 0.87. Receiver operating characteristic curves were generated for the visual and computer ST depression in lead V5. Analysis of the areas under these curves showed no significant difference between them, indicating that computer-assisted analysis was not superior to unmodified visual analysis. A similar analysis was applied to two other computer indexes reported to be superior to visual assessments (treadmill exercise score and ST index). These computer indexes were not superior to a conventional visual analysis of leads I, II, V2, V4 and V5 in predicting severe disease (greater than 50% luminal narrowing). These results suggest that computer-assisted interpretation does not improve the accuracy of exercise electrocardiography in diagnosing coronary artery disease in subjects without prior myocardial infarction.