Exercise tomographic thallium-201 imaging in patients with severe coronary artery disease and normal electrocardiograms

Objective: To assess the incremental value and cost-effectiveness of exercise tomographic thallium-201 imaging compared with clinical and exercise electrocardiographic variables for detecting three-vessel or left main coronary artery disease in patients with normal at-rest electrocardiograms. Design: Prospective cohort study. Participants: 411 patients (77 [19%] had three-vessel or left main disease) with normal at-rest electrocardiograms who underwent exercise tomographic thallium-201 studies and subsequently had coronary angiography. Measurements: Clinical, exercise, and thallium-201 variables; univariate followed by multivariate logistic regression analysis to determine predictors of three-vessel or left main disease (clinical variables; clinical and exercise electrocardiographic variables; and clinical, exercise, and thallium-201 variables). Patients were classified by each of these models into low-, intermediate-, and high-risk groups. Setting: A tertiary referral center. Results: Among the clinical variables, diabetes mellitus, sex, age, and typical angina were independently associated with severe coronary disease (46% of patients were correctly classified into low- or high-risk groups). The peak exercise heart rate-blood pressure product and the magnitude of the exercise-induced ST depression added independent information to clinical variables. Among the thallium variables, the change in the global thallium-201 score (a measure ot redistribution) added independent information to clinical and exercise variables, resulting in only 3% of the patients being reclassified regarding their predicted risk for severe coronary disease. The cost per additional reclassification was estimated to be $20 550. Twenty-one cardiac events occurred (7 cardiac deaths and 14 myocardial infarctions) after thallium study (follow-up, 2.8±1.0 years). Event-free survival was 94% to 97% regardless of the predicted probability of developing three-vessel or left main coronary artery disease by any model. Conclusions: When the at-rest electrocardiogram is normal, thallium-201 scintigraphy adds little information to clinical and exercise variables in identifying patients with severe coronary artery disease. The high cost of this information may not justify the routine use of sophisticated imaging for this purpose.