To assess the effect of heart rate adjustment of the magnitude of the ST integral (ST-HR integral) on exercise test performance, the exercise electrocardiogram (ECG) of 50 clinically normal subjects and 100 patients with known or suspected coronary artery disease was analyzed. At matched specificity of 96% with standard ECG criteria (greater than or equal to 0.1 mV of additional horizontal or downsloping ST segment depression), an unadjusted ST integral partition of 16 microV-s identified coronary disease in the 100 patients with known or suspected disease with a sensitivity of only 41%, a value significantly lower than the 59% sensitivity of standard ECG criteria (p less than 0.01) and the 65% sensitivity of an ST depression partition of 130 microV (p less than 0.001). However, test performance of the ST integral was greatly improved by simple heart rate adjustment: at a matched specificity of 96%, an ST-HR integral partition of 0.154 microV-s/beat per min identified coronary disease in the 100 patients with a sensitivity of 90%, a value significantly greater than the 59% sensitivity of standard criteria and 65% sensitivity of ST depression criteria (each p less than 0.001) and similar to the 91% sensitivity of the ST-HR index and 93% sensitivity of the ST-HR slope (each p = NS). Comparison of receiver-operating characteristic curves confirmed the superior overall test performance of the ST-HR integral relative to the ST integral and ST segment depression, and demonstrated improved performance that was comparable with that of the ST-HR index and the ST-HR slope.(ABSTRACT TRUNCATED AT 250 WORDS)