BACKGROUND: Risk assessment may be important in patients being considered for repeat revascularization after prior coronary intervention or surgery. We sought the prognostic value of radionuclide stress myocardial perfusion imaging or echocardiography among patients with previous revascularization. METHODS: Studies on the outcomes of stress imaging tests after revascularization were selected from an electronic search if they reported the odds or hazard ratio (HR) of an abnormal stress test in the prediction of mortality (cardiac or total), hard cardiac events (cardiac death and myocardial infarction [MI]), total hard events (total mortality and MI]), or overall events (cardiac death, MI, and repeat revascularization). RESULTS: In 29 studies (12,874 patients, 63 ± 3 years, 80% men), an abnormal test result was associated with hard cardiac events (HR 1.2, 95% CI 1.1-1.3), cardiac mortality (HR 5.8, 95% CI 0.8-10.8), total mortality (HR 2.2, 95% CI 1.3-3.1), total hard events(HR 2.4, 95% CI 1.4-3.3), and overall events (HR 1.2, 95% CI 1.1-1.3). The nature of the end point was not associated with differences in the prediction of events, but the type of revascularization showed a significant association with outcome, with percutaneous intervention portending a worse outcome. Age and the timing of the stress imaging postrevascularization were inversely associated with survival. Gender, length of follow-up after testing, symptom status, past infarction, and risk factor status did not explain interstudy heterogeneity. CONCLUSIONS: In patients with previous revascularization, abnormal results at stress echocardiography or radionuclide myocardial perfusion imaging are predictive of subsequent events, with age, type of revascularization, and the timing of the stress imaging after revascularization being important sources of heterogeneity between studies.