Multislice computed tomography for determination of coronary artery disease in a symptomatic patient population

Multislice computed tomography (MSCT) has started to replace Electron beam CT for quantitation of coronary artery calcium. However no study has evaluated the diagnostic accuracy of MSCT for prediction of coronary artery disease (CAD) in a symptomatic patient population using the volume score.

1347 symptomatic subjects (male = 803, mean age = 62 years) with suspected CAD underwent MSCT studies 1 +/- 2 days before the coronary angiogram. The Agatston (ACS) and Volumetric calcium score (VCS) were calculated using a proprietary workstation. Statistical analyses included the Pearson's correlation coefficient and the nonparametric Mann-Whitney U-test to compare the calcium score in different age groups and between men and women. Sensitivity, specificity and predictive accuracy were calculated for different calcium thresholds for prediction of CAD. ROC curve analyses were used to establish relations between the coronary calcium score and presence or absence of CAD. In 720 (53%) subjects (male = 419) angiography revealed a minimal lumen diameter stenosis greater than 50%. Patients with significant CAD had significantly higher total calcium score values than patients without CAD (P = 0.001). ACS and VCS demonstrate a close correlation for the whole study group, r = 0.99. The overall sensitivity of any calcium to predict stenosis was 99%, specificity = 32%. Exclusion of calcium was highly accurate for exclusion of CAD in subjects older than 50 years (predictive accuracy = 98%). An absolute cutoff >100 and an age and sex specific threshold (score over 75th percentile) were identified as the cutoff levels with the highest sensitivities (86-89%) and lowest false positive rates (20-22%). ROC analyses revealed MSCT calcium scanning as a good clinical test which can be performed with similar accuracy in all age groups with an area under the curve of 0.84.

Determination of coronary calcium with MSCT is an accurate imaging modality for prediction of significant CAD in a patient population with intermediate likelihood of CAD. Exclusion of any calcium provided strong evidence that patients older than 50 years did not have obstructive CAD. ACS and VCS show an equivalent diagnostic accuracy.