RESUMO
Objectives: To determine the diagnostic performance and influencing factors of 128-slice coronary computed tomography angiography (CCTA) compared with invasive coronary angiography (ICA) in patients with suspected coronary artery disease (CAD). Methods: A cross-sectional analysis study enrolled 139 patients suspected of having CAD, who underwent and received a 128-slice CCTA and ICA. Results: The patient-based model showed high sensitivity and a positive predictive value of 93.2% and 95.3%, respectively (for stenosis ≥50%). However, these values were lower when analyzed using vessel-based (85.6% and 81.1%) and segment-based (73.9% and 66.6%) models. Specificity and negative predictive value were highest in the segment-based model, decreasing in vessel- and patient-based models at 96.4% and 95.4%, 90.5% and 90.0%, and 36.4% and 42.1%, respectively (for stenosis ≥70%). All diagnostic values were reduced when the calcium score was ≥400 Agatston units. Conclusion: 128-slice CCTA is an optimal, minimally invasive, and high-performance method to diagnose the stenosis and morphology of coronary artery lesions. The diagnostic performance of 128-slice CCTA is very high. Heart rate and body mass index do not affect diagnostic accuracy, whereas a calcium score ≥400 Agatston units is a factor that causes a decrease in diagnostic performance.
