Coronary artery disease (CAD) is the leading cause of morbidity and mortality in patients with diabetes mellitus. In fact, patients with diabetes have the same risk of myocardial infarction as do nondiabetic subjects with a history of infarction. For this reason, diabetes has been designated by the American College of Cardiology (ACC) and the American Heart Association (AHA) as a CAD equivalent. For women, data indicate a substantially elevated risk of cardiovascular disease (CVD) even before a clinical diagnosis of type 2 diabetes has been made. Identifying patients with diabetes who have CAD and who will benefit from medical and/or invasive intervention to prevent cardiovascular events is a challenge in both symptomatic and asymptomatic patients. The decision to evaluate patients with diabetes who are asymptomatic for CAD presents the greatest challenge; investigation will reveal 10% to 15% of these patients to have CAD. Current diagnostic tools include exercise tolerance testing, stress echocardiography, stress myocardial perfusion imaging (MPI), and cardiac catheterization. Few guidelines are available to aid in the choice of testing modalities for a given patient. Although cardiac catheterization is useful, it is generally reserved for patients in whom invasive intervention is suitable. The American Diabetes Association (ADA) recommends exercise tolerance testing alone in symptomatic patients with > or = 2 CAD risk factors or an abnormal resting electrocardiogram (ECG). However, that recommendation is not based on data; it is the consensus of an expert panel. Stress echocardiography is a useful, noninvasive procedure; however, there is limited experience with this technology in the diabetic population. Recently accumulated data support both diagnostic and prognostic roles for stress MPI, particularly with ECG-gated single-photon emission computed tomographic imaging. In symptomatic patients with diabetes, the presence and extent of abnormal stress MPI findings have been found to be highly accurate independent predictors of subsequent cardiac events: 18% to 26% of asymptomatic patients with diabetes have perfusion defects consistent with CAD. However, CVD risk factors are not predictive of abnormal MPI findings even though duration of diabetes and abnormal ECGs are. The results of future studies may be helpful in guiding the selection of asymptomatic patients to undergo myocardial perfusion and function studies. In conclusion, MPI provides clinicians with an important diagnostic tool, because it offers perfusion as well as functional information for diagnosis and risk stratification in patients with diabetes. These capabilities facilitate decision making regarding the appropriateness of medical therapy or surgical intervention in these individuals.