Technetium-99m (Tc-99m)-tetrofosmin is a radio isotope that has been shown to be an accurate alternative to thallium-201 for detecting coronary artery disease. However, its prognostic value is less well determined. To this end, 459 consecutive patients (mean age 58 +/- 10 years) with suspected or known coronary artery disease underwent exercise single-photon emission tomography Tc-99m-tetrofosmin scintigraphy. Follow-up, defined as the time from scanning until a soft event (revascularization procedures), a hard event (myocardial infarction and cardiac death), or patient response, lasted up to 78 months (median 38). An ischemic scintigraphic perfusion score, which takes into account both the extent and severity of reversible perfusion defects, was calculated to estimate the severity of perfusion abnormalities. Patients with normal scans were at low risk of events (yearly hard event rate 0.5% and soft event rate 0.9%). The rate of outcomes increased significantly with abnormal scans (yearly hard event rate 4.9% and soft event rate 10.3%). Statistical analysis using the Kaplan-Meyer survival curves showed a significant difference in event-free survival between patients with normal and abnormal scans. With use of Cox proportional-hazards analysis, after adjusting for prescan information, nuclear data provided incremental prognostic value for hard events (clinical and exercise data vs nuclear data; chi-square = 15.5 vs 33.4, p <0.001). Exercise single-photon emission tomographic scintigraphy using Tc-99m-tetrofosmin provides significant independent information on the subsequent risk of hard and soft events. The annual event rate for hard and soft events is <1% for patients with a normal scan. Furthermore, this tracer yields incremental prognostic information in addition to that provided by clinical and exercise data for hard events.