Exercise testing provides additional prognostic information in angiographically defined subgroups of patients with coronary artery disease.

We examined whether exercise testing with measurement of cardiac output during maximal exercise can provide additional prognostic information for medically treated patients in whom left ventricular function and extent of coronary artery disease are known. We followed 1034 patients with normal or mildly impaired left ventricular function; 410 of these patients (group 1) had single-vessel disease, 316 had double-vessel disease (group 2), and 308 had triple-vessel disease (group 3). In addition, 204 patients with double- or triple-vessel disease and moderately impaired left ventricular function (group 4) were followed. Mean follow-up in these 1238 patients was 4.5 years. End point of follow-up was death. Groups 1, 2, and 3 were divided into terciles according to the maximally achieved values of the following exercise variables: exercise tolerance, angina-free exercise tolerance, maximal heart rate, and cardiac output during maximal exercise. Group 4 was divided into halves accordingly. Survival curves (according to the method of Cutler and Ederer) for group 2 showed a 15% difference in 5 year survival rate between the highest and lowest terciles (p less than .005) by use of the noninvasive variables exercise tolerance, angina-free exercise tolerance, and maximal heart rate (95% vs 80%). The separation into terciles according to cardiac output during maximal exercise resulted in a significant difference in survival rates between the highest and lowest terciles (halves) in all groups of patients. The differences in 5 year survival rates were 9% (p less than .05), 16% (p less than .05), and 19% (p less than .005) for groups 1, 2, and 3, respectively, and 22% for group 4 (p less than .005).(ABSTRACT TRUNCATED AT 250 WORDS)