Objectives. The objective of this study was to examine the independent and incremental prognostic value of exercise singlephoton emission computed tomographic (SPECT) thallium imaging in patients with angiographically defined coronary artery disease. Background. Previous studies showed the importance of exercise thallium-201 in risk stratification. However, most of these studies used planar imaging techniques. Methods. Follow-up data were obtained in 316 medically treated patients with coronary artery disease. Cox proportional hazards regression models were used to examine the independent and incremental prognostic values of clinical exercise, thallium and cardiac catheterization data. Results. There were 35 events (cardiac death or nonfatal myocardial infarction) at a mean follow-up time of 28 months. Univariate analysis showed that gender (chi-square = 5.1), exercise work load (chi-square = 3.1), extent of coronary artery disease and left ventricular ejection fraction (chi-square = 14.8) and thallium variables (chi-square = 22.7) were prognostically important. The thallium data provided incremental prognostic value to catheterization data (chi-square = 33.7, p < 0.01). The extent of the perfusion abnormality was the single best predictor of prognosis (chi-square = 14). Patients with a large perfusion abnormality had a worse prognosis than that of patients with a mild or no abnormality (Mantel-Cox statistics = 10.6, p < 0.001). Conclusions. In medically treated patients with coronary artery disease, exercise SPECT thallium imaging provides independent and incremental prognostic information even when catheterization data are available. The extent of the perfusion abnormality is the single most important prognostic predictor. © 1993.