Accuracy of 64-multidetector-row computed tomography in the diagnosis of coronary artery disease

Abstract

Objectives: The aim of this prospective study was to assess the accuracy of 64-multidetector-row computed tomography coronary angiography (CTA) in the diagnosis of coronary artery disease (CAD).

Patients and methods: Ninety-two patients suspected of having CAD underwent CTA using a 64-slice CT scanner before a scheduled, conventional coronary angiogram (CCA). Blinded assessment of CTA to detect CAD was performed. The accuracy of CTA in detecting significant stenoses (> or =50%) was compared to CCA. Data analysis was performed on 73 patients because the scans were nondiagnostic in 5 patients and 14 refused to undergo coronary angiography.

Results: The CTAs of 21 of these 73 patients were considered as normal; 19 were confirmed on CCA. For the remaining 52 diagnosed as abnormal, 51 were confirmed on CCA. For patient-based analysis, CTA had a sensitivity of 95%, a specificity of 96%, a positive predictive value of 98% and a negative predictive value of 90%. For the whole vessel, the sensitivity of CTA was 60-100%, for all vessels and the specificity was 82-100%. Pooled sensitivity was 92% and pooled specificity was 98%. For the segments, the sensitivity of CTA was 64% or above for all vessels except for the distal left anterior descending artery (40%), mid circumflex artery (50%) and posterior descending artery (60%); the pooled sensitivity was 79%. The specificity for the segments was 82-100% for all vessels and pooled specificity was 94%.

Conclusion: The sensitivity and specificity for patient-based analysis and for the main coronary vessels were high whereas for the segments, the sensitivity was moderately good, but the specificity was high, confirming that a negative CTA is useful to rule out significant CAD. A coordinated classification system between radiologists and cardiologists is required to eliminate errors in segment classification.