The importance of clinical subsets in interpreting maximal treadmill exercise test results: the role of multiple-lead ECG systems.

1 day before their coronary arteriogram. The prevalence of coronary stenoses 2 70% was 86% in 87 men with typical angina, 65% in 64 men with probable angina, and 28% in 49 men with nonspecific chest pain (p < 0.001). Among the 117 men who had coronary disease and probable or typical angina, there was more three-vessel (32% vs 7%;p = 0.06) and proximal left coronary disease (41% vs 14%;p < 0.05) than among the 14 men with coronary lesions and nonspecific chest pain. Prevalence, extent and location of coronary stenoses, therefore, were different among the clinical subsets. The predictive value of a positive test in any one of 14 leads was 45% (nine of 20) in men with nonspecific chest pain vs 82% (36 of 44) in men with probable angina and 100% (65 of 65) in men with typical angina (p < 0.001). The predictive value of a negative test in 14 leads was 83% (24 of 29) in men with nonspecific chest pain vs 70% (14 of 20) in men with probable angina and 55% (12 of 22) in men with typical angina (NS). In men with probable or typical angina, 92% (33 of 36) of those with a positive test and treadmill work time 360 seconds had multivessel disease; only one man in 40 with a negative test in 14 leads and treadmill work time > 540 seconds had three-vessel disease. The diagnostic impact of maximal treadmill testing using 14 ECG leads is greatest in men with typical and probable angina. In these two clinical subsets the presence or absence of horizontal or downsloping ST-segment depression 1 mm, ST-segment elevation > 1 mm, or a slowly upsloping ST-segment depression > 2