The Predictive Value of Vessel-Based Calcium Score in the Detection of Coronary Stenosis

Abstract

The aim of this study was to investigate the predictive value of vessel-based calcium (Ca) score in the detection of >=%50 coronary stenosis. The results of 65 patients with typical chest pain or multiple cardiovascular risk factors who underwent coronary Ca scoring and catheter angiography were retrospectively evaluated. 260 coronary arteries were analysed and mean vessel-based Ca score was 179 +/- 17.0 (range 0-1729). In 114 coronary arteries with vessel-based Ca scores of >= 100, >=%50 stenosis was found in 90 of them (79%). On the other hand, in 146 coronary arteries with vessel-based Ca scores of <= 99, >=%50 stenosis was found in 17 of them (12%). Sensitivity, specificity, positive predictive value, negative predictive value and accuracy results for different vessel-based Ca scores in the detection of >=%50 coronary stenosis were as follows: 84%, 84%, 79%, 88%, 84% respectively for the cut-off value of 100; 39%, 99%, 95%, 70%, 74% respectively for the cut-off value of 400; 7%, 100%, 100%, 60%, 62% respectively for the cut-off value of 1000. The cut-off value of the vessel-based Ca score was 100 in the ROC curve analysis. During Ca scoring, calculation of both total Ca score and vessel-based Ca score will increase the diagnostic accuracy of coronary artery stenosis. A cut-off value of 100 for the vessel-based Ca score seems to be good in the prediction of >= 50% coronary stenosis.