This new article by Nikolaou K et al in the July issue of the AJR is one more article that shows the significance of this technique. In their results, "Sensitivity, specificity, and the negative predictive value (NPV) of 64-MDCT per patient were 97%, 79%, and 96%, respectively".
As with all similar articles, the negative predictive value was very high. The specificity was low but to maintain a high sensitivity, this is acceptable. The main reason for the low specificity was the overestimation of disease. In fact the logic that they have used and which makes sense practically is that the idea is not to miss a single patient of coronary artery disease (CAD). If that means that a few lesions are overdiagnosed, that is a much better evil than to underdiagnosis potentially significant disease. Also, once one significant lesion is diagnosed in a patient and it is obvious, he/she needs to go for a catheter angiography, the need to be very accurate in the rest of the vessels is less of an issue as well.
This is a 56-years old doctor who had an attack of angina. He came for a cardiac CT and we found a severe distal LAD stenosis (Fig. 1), which was confirmed on angiography (Fig. 2). There was another lesion in the proximal RCA (Fig. 3), which was overdiagnosed as compared to the catheter angiogram (Fig. 3). The proximal PDA lesion was well seen on both studies (Figs. 5, 6).