Background: The sensitivity of coronary calcification by electron beam tomography (EBT) for the detection of coronary atherosclerosis is well-established. Combining the anatomic information of EBT with the functional information of stress testing might reduce the high false-positive result rate seen with treadmill stress test (TMST) alone. No studies have reported the additive value of a negative EBT result (no coronary calcium) for excluding obstructive coronary artery disease (OCAD) in patients with a positive TMST result. This study evaluated the negative predictive value and potential clinical utility of EBT to identify patients with a falsely abnormal TMST.
Methods: A coronary calcium score was determined by EBT for 153 symptomatic patients who underwent coronary angiography because of a positive TMST. The sensitivity, specificity, and predictive values of EBT were determined. A multiple logistic regression analysis compared conventional cardiac risk factors with coronary calcification by EBT for predicting OCAD. A receiver operating characteristic curve was generated plotting sensitivity versus false-positive rate.
Results: The false-positive rate of the TMST compared with angiography was 27% (41 of 153). The sensitivity of a nonzero coronary calcium score for OCAD was 98% (110 of 112), with a negative predictive value of 93%. According to multiple logistic regression, coronary calcification by EBT was a stronger predictor of OCAD than any conventional cardiac risk factor. EBT calcium added incremental value to the results of the TMST. Receiver operating characteristic curve analysis found an area under the curve of 0.91 (P <.001).
Conclusions: The absence of coronary calcification by EBT reliably identified patients with a false-positive TMST result. The combination of EBT with TMST is a potentially useful diagnostic strategy to reduce the number of false-positive test results.