Discrepant findings of computed tomography quantification of minimal lumen area of coronary artery stenosis: Correlation with intravascular ultrasound

Objectives: To study the diagnostic performance of computed tomography (CT) quantification of minimal lumen area (MLA) based on multiple factors (image quality, calcification and lesion locations), with reference to intravascular ultrasound (IVUS). Methods: Consecutive clinically ordered patients were prospectively enrolled in our study. CT quantification of MLA was manually measured on cross-sectional view and further compared with IVUS findings. A significant lesion was defined as <= 6 mm(2) MLA for the left main (LM) coronary artery and a <= 4 mm(2) MLA for other epicardial vessels. Results: Non-calcified lesions had good correlation between CT and IVUS (r = 0.96) and the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were 85% (51/60), 93.94% (31/33), 96.23% (51/53) and 77.5% (31/40) respectively. Decreased correlation (r = 0.814) lower specificity (23.81% (5/21)) and positive predictive value (60.98% (25/41)) were observed in calcified subgroup. Artifact-absent lesions demonstrated excellent correlation (r = 0.967) and the sensitivity, specificity, PPV and NPV were 96% (48/50), 93.94% (31/33), 96% (48/50) and 93.94% (31/33) respectively. Impaired correlation (r = 0.584) was noted in artifact-present lesions and the sensitivity, specificity, PPV and NPV were 80% (28/35), 23.81% (5/21), 63.64% (28/44) and 41.67% (5/12) respectively. Excellent correlation between CT and IVUS was noted in proximal epicardial vessels (r = 0.908) and the sensitivity, specificity, PPV and NPV were 92.31% (36/39), 78.13% (25/32), 83.72% (36/43) and 89.29% (25/28) respectively. Middle right coronary artery lesions showed non-significant correlation (r = 0.54, p = 0.055). Conclusions: CT quantification of MLA can only be accurately achieved in non-calcified lesions with absence of artifact. Diagnostic performance is impaired in calcified lesions. (C) 2012 Elsevier Ireland Ltd. All rights reserved.