Background: The incremental diagnostic value of dobutamine stress echocardiography (DSE) and 99mTc-labeled sestamibi single-photon emission computed tomography for the evaluation of the presence and extent of coronary artery disease (CAD) was assessed with ordered logistic regression and receiver-operating characteristic curves.
Methods and results: Forty-five consecutive patients (33 men and 12 women; 53 +/- 6.8 years) underwent exercise electrocardiography and simultaneous DSE and sestamibi single-photon emission computed tomographic imaging. Coronary angiography was performed in all patients (significant coronary stenosis > 50%). On the basis of the results of exercise electrocardiography, the pretest probability for CAD (Diamond's algorithm) was low (45.6% +/- 12.7%). According to ordered logistic regression analysis, some models were estimated that performed a diagnostic accuracy level for CAD. In particular, we evaluated a clinical model (model 1) determined by the following parameters: sex, age, presence of chest pain, and positivity of electrocardiogram during dobutamine stress test. This model was 64.3% +/- 10.7% accurate for the prediction of CAD. The addition to model 1 of DSE parameters (wall motion stress and rest score index and relative difference) (model 2) yielded a diagnostic accuracy of 81.4% +/- 4.3% (p < 0.045), whereas the addition to model 1 of single-photon emission computed tomographic parameters (the difference between perfusional stress and rest score index) (model 3) improved diagnostic accuracy to 92.3% +/- 5.5% (p < 0.003), a level that appeared significantly higher than that of model 2 (p < 0.016).
Conclusion: Both noninvasive methods for the detection of CAD showed a good diagnostic accuracy, especially when test-derived parameters were combined with clinical data. Nevertheless, the single-photon emission computed tomographic model showed a higher sensitivity compared with the DSE model.