Treadmill exercise electrocardiography (ECG) is one of the most common noninvasive methods for detecting ischemic heart disease. However, this method has problems due to false-positive and false-negative results in a significant number of patients. The aim of this study was to determine whether the diagnostic accuracy of treadmill exercise ECG for detecting significant coronary stenosis can be improved by employing a step-up diagnostic method using multiple diagnostic indicators. We studied 273 consecutive patients (mean age, 56 +/- 9 years; 190 men and 83 women) without a history of myocardial infarction who underwent treadmill exercise ECG and coronary angiography for ischemic chest pain. Of these, 146 patients had no significant coronary stenosis, 61 had single-vessel disease, 56 had multivessel disease, and 10 patients had left main truncus disease. A multivariate logistic regression analysis was used to select 3 treadmill exercise electrocardiographic parameters that were independent predictors of the presence or absence of significant coronary stenosis: exercise-induced maximum ST-segment depression, QT dispersion immediately after exercise, and Athens QRS score. Significant coronary stenosis was diagnosed with a sensitivity of 84% and a specificity of 90% when a step-up diagnostic method using these 3 indicators was employed. These results were better than those obtained for each indicator alone (exercise-induced maximum ST-segment depression: sensitivity, 66%, and specificity, 73%; QT dispersion immediately after exercise [> or =60 ms positive]: sensitivity, 76%, and specificity, 86%; and Athens QRS score [< or =5 mm positive]: sensitivity, 72%, and specificity, 72%). We conclude that this step-up diagnostic method, using multiple diagnostic indicators, is a clinically useful predictor of the presence or absence of significant coronary stenosis.