Abstract 6104: Identification and Characterization of Severe Coronary Artery Stenoses by Dual-Source 64-Slice Coronary Computed Tomography Angiography: Comparison with Invasive Coronary Angiography
Percutaneous coronary intervention (PCI) of complex coronary lesions requires a high level of procedural expertise. Knowledge of such lesions before invasive coronary angiograph (ICA) may streamline PCI performance. We evaluated the accuracy of CCTA in detecting severe coronary stenoses (SCS) and in describing SCS as high risk for PCI. We studied all non-bypassed, non-stented, ≥2mm diameter native vessel segments in 85 consecutive patients (pts) who underwent 64-slice CCTA (Siemens, dual-source) followed by ICA within 30 days. Two CCTA readers qualitatively and quantitatively identified all SCS, defined as causing ≥70% visual diameter obstruction, and characterized each SCS as type-C if any morphologic feature met current ACC criteria for high PCI risk (ostial or major side branch involvement, total occlusion, >20mm length, tortuous proximal vessel, >90° bend at lesion). Results were compared to ICA images similarly analyzed by 2 blinded interventional cardiologists. PCI time and contrast use were compared between pts with and without type-C SCS on CCTA. Mean calcium score was 734. ICA found 93 SCS in 52 pts. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for CCTA in detecting SCS were: by segment (n=940), 90%, 98%, 83%, and 99%; by artery (n=328), 85%, 95%, 85%, and 98%; by pt (n=85), 94%, 76%, 86%, 89%. Among 84 SCS correctly identified by CCTA, sensitivity, specificity, PPV, and NPV for type-C lesions were: by segment (n=84), 86%, 80%, 86%, and 80%; by artery (n=73), 83%, 81%, 89%, and 72%; by pt (n=49) 88%, 87%, 94%, and 77%. The most frequent false negative type-C finding on CCTA was lesion length >20mm (8 cases); the most frequent false positive was major side branch involvement (12 cases). Presence of type-C morphology on CCTA was independently and strongly associated with longer PCI time (42.63 ± 24.67 min vs. 21.2 ± 13.26 min) and increased contrast use (262.86 ± 149.96 ml vs. 139.43 ± 47.38 ml). In a population with a high burden of coronary artery disease, 64-slice CCTA capably detects SCS and SCS morphology that increase PCI difficulty. CCTA can be helpful in identifying pts with SCS who significantly benefit from expert consultation prior to or at the time of PCI.