We examined 1,159 consecutive patients who underwent adenosine stress dual isotope single-photon emission computed tomography (SPECT) and had follow-up performed at a mean of 27.5 +/- 9.1 months (94% complete) for hard events (cardiac death and myocardial infarction) and referral to cardiac catheterization after nuclear testing. During follow-up, 120 hard events occurred (11.0% hard event rate; 72 cardiac deaths [6.7% cardiac death rate] and 57 myocardial infarctions [5.3% myocardial infarction rate]). Cox proportional hazards analysis revealed that nuclear testing added incremental value after adjusting for clinical and historical variables (global chi-square increased 13 to 98 for cardiac death as the end point, global chi-square increased 19 to 105 for hard events as the end point; p <0.0001 for both). Kaplan-Meier analysis demonstrated that after clinical risk stratification of the patient population, the results of nuclear testing were further able to significantly stratify both low- and intermediate- to high-risk patients. Patients with both normal and mildly abnormal scans were at low risk of cardiac death (<1% cardiac death per year of follow-up) and the risk of events increased significantly with worsening scan result. Multivariable analysis revealed that the only predictor of referral to catheterization was the extent and severity of reversible defect present on the scan. Referral rates to early catheterization were very low in patients with normal scans and increased significantly as a function of worsening scan results. In patients who underwent myocardial perfusion SPECT using adenosine stress, the results of nuclear testing yielded incremental prognostic information and clinically relevant risk stratification. Referring physicians predominantly utilized nuclear information when referring patients to catheterization after nuclear testing and do so at rates comparable with those after exercise SPECT despite the higher risk of events in patients undergoing pharmacologic stress.