Myocardial imaging with 201thallium: an analysis of clinical usefulness based on Bayes' theorem.

Rest-exercise thallium-201 (201Tl) myocardial imaging and rest-exercise electrocardiography were performed in 137 patients with suspected coronary artery disease (CAD). The final diagnosis of coronary disease was made by arteriography. Sensitivity and specificity for the ECG and thallium studies alone or combined were then determined. Based on these data, the posttest probability of CAD with a normal or abnormal test was calculated using Bayes' theorem for disease prevalences ranging from 1%--99%. The difference between the probability of disease with a normal test and the probability of disease with an abnormal test was also calculated for each prevalence range. The results demonstrate that 201Tl imaging discriminates between disease absence or presence better than does the ECG. However, both the ECG and thallium studies provide rather poor discrimination between disease and no disease when the disease prevalence is low (less than 0.20) or high (greater than 0.70). Because of this characteristic, it is unlikely that screening tests for CAD will prove useful unless the disease prevalence in the group under study is in the moderate (0.20--0.70) range.