Comparison of technetium-99m sestamibi left ventricular wall motion and perfusion studies with thallium-201 perfusion imaging: in search of the combination of variables with the highest accuracy in predicting coronary artery disease.

Measurements of myocardial perfusion and ventricular function are expected to provide additional information in the detection of coronary artery disease (CAD). The purpose of this study was threefold: (1) to determine to what extent technetium-99m sestamibi wall motion yields different information compared with 99mTc-sestamibi and thallium-201 perfusion; (2) to test which information unique to either study is of value in diagnosing CAD; and (3) to assess the combination of variables with the highest diagnostic accuracy. Perfusion and wall motion scores (at rest and during exercise) obtained from visual and quantitative planar 201Tl and 99mTc-sestamibi scintigraphy of 60 patients with suspected CAD were compared with the angiographic results by means of a polytomous logistic regression model and the diagnostic values were compared with one another. All univariate variables were significantly related to the probability of CAD and its extent. Comparative studies revealed a large degree of correlation between 201Tl stress and redistribution variables. The rest 99mTc-sestamibi and wall motion studies contained partially different information. Stepwise logistic regression analysis showed the strongest diagnostic power for the combination of 201Tl visual analysis of the stress images with quantitative redistribution images (sensitivity 93%, specificity 71%). The diagnostic power was similar for all combinations of visual and quantitative analyses of the exercise and redistribution images. The strongest diagnostic power of the 99mTc-sestamibi variables was the score of the diastolic stress image (sensitivity 91%, specificity 79%). Comparable sensitivity and specificity estimates were found when both optimal models were compared. Wall motion studies did not have additional diagnostic power. Although 99mTc-sestamibi wall motion studies, both at rest and during exercise, provide information in addition to the 99mTc-sestamibi or 201Tl myocardial perfusion variables, the information does not enhance the diagnostic power with regard to the prediction of CAD.