Assessment of prognosis plays an important role in the management of patients with CAD. The objective of the study was to improve risk stratification in patients with known coronary angiographic findings. We analyzed the prognostic importance of 13 angiographic, exercise, and clinical variables in 1183 medically treated patients with documented CAD. Five-year actuarial survival rate (5-YSR) was 87%. Multivariate analysis with the proportional hazards regression model revealed four continuous and one discrete variable to be of independent prognostic importance (chi 2 value): cardiac output at the highest work load (COmax) (chi 2 = 80.7); coronary score (chi 2 = 18.6); heart volume by X-ray (chi 2 = 14.7); maximal pulmonary wedge pressure during exercise (chi 2 = 5.3), and history of myocardial infarction (chi 2 = 4.8). Inclusion of these variables in the survival function according to the regression model resulted in excellent prediction of 5-YSR, e.g. in the patients with three-vessel disease (N = 399): actuarial 5-YSR was 80%, calculated 81%. Patients with three-vessel disease and COmax greater than 11.21 min-1 (N = 188) had an actuarial 5-YSR of 88%, calculated 89%; if COmax was less than 11.21 min-1 the actuarial 5-YSR was 71%, calculated 70%. Patients with three-vessel disease from an independent cohort of surgically treated patients (N = 507) had a calculated 5-YSR under an assumed medical regimen of 77%.(ABSTRACT TRUNCATED AT 250 WORDS)