Prediction of severe coronary artery disease and long-term outcome in patients undergoing vasodilator SPECT

Background. Vasodilator perfusion imaging has not been extensively evaluated for predicting severe coronary artery disease (CAD) or long-term prognosis. Methods and Results. The goals of this study were to develop a model to predict left main/3-vessel CAD in patients undergoing vasodilator thallium 201 imaging and coronary angiography (angiographic population) and to test the long-term prognostic value of this model in a separate cohort of patients who were not referred for angiography (prognostic population). In the angiographic population (n = 653) the X² value of the clinical model (containing the variables age, sex, and prior myocardial infarction) in the prediction of severe CAD was 32. The addition of 3 vasodilator Tl-201 variables (magnitude of ST-segment depression, summed reversibility score, and increased lung uptake) increased the model X² value to 114 (P < .001). Only 9% of predicted low-risk patients versus 57% of predicted high-risk patients had severe CAD. In the prognostic population (n = 521) survival rates free of cardiac death or myocardial infarction at 7 years were 91%, 73%, and 51%, respectively, for patient groups predicted to be at low, intermediate, and high risk of severe CAD (P < .001). Conclusions. Clinical and vasodilator Tl-201 variables can accurately predict the risk of severe CAD. Stress Tl-201 variables add incremental information to clinical variables. The same model also predicts patients outcome.