Abstract
Background and Objectives: Coronary computed tomography angiography (CCTA) is widely used in the diagnostic evaluation of suspected stable coronary artery disease; however, its agreement with invasive coronary angiography (ICA) remains inconsistent across different levels of analysis. The aim of this study was to evaluate the agreement between CCTA and ICA and to identify the factors associated with discrepancies. Materials and Methods: A single-centre retrospective analysis of 500 patients was performed. All patients underwent CCTA within one year prior to ICA. Coronary stenoses were evaluated at the 11-segment coronary artery, vessel, and patient levels using a ≥50% cut-off. Diagnostic agreement was assessed using the kappa coefficient, while diagnostic performance was evaluated in terms of sensitivity, specificity, positive predictive value, and negative predictive value. Factors associated with discrepancies were evaluated using a logistic regression model. Results: At the segment level, agreement between CCTA and ICA was low to moderate across 11 coronary segments (κ = 0.108-0.461). At the patient level, CCTA identified ≥50% coronary stenosis more frequently than ICA (86.2% vs. 59.4%, p < 0.001), demonstrating high sensitivity (91.3%) but low specificity (21.2%). Diagnostic discrepancies were associated with higher coronary calcium burden, and in multivariable analysis, body mass index > 25 kg/m(2), age < 68 years, and multiple comorbidities were independently associated with discordant findings. Conclusions: At the patient level, CCTA demonstrates high sensitivity and represents an appropriate non-invasive method for patient selection for further diagnostic evaluation. However, agreement between CCTA and invasive coronary angiography remains limited at the segment and vessel levels. Diagnostic discrepancies were significantly associated with coronary artery calcification and higher body mass index (BMI), which should be taken into consideration when interpreting CCTA findings.