Improved evaluation of coronary artery diseases from patients with coronary calcification utilizing FFRCT: a comparative study against CCTA.

Abstract

Objective: This multicenter study evaluated the diagnostic accuracy of coronary computed tomography angiography (CCTA) and computed tomography-derived fractional flow reserve (FFR_CT) in diagnosing coronary artery disease (CAD), focusing on the impact of coronary calcification, using invasive coronary angiography (ICA) as the reference.

Methods: The study analyzed 4172 patients from 3 centers who underwent CCTA and FFR_CT between August 2021 and August 2022. Exclusion criteria included the absence of ICA within 90 days after CCTA, left main disease, previous coronary revascularization, or unmeasurable Agatston score (AS).

Results: The study included 492 patients. In patients with less than mild calcification (AS < 100), the area under the receiver operating characteristic curve (AUC) for FFR_CT was superior to CCTA (0.87 [95% confidence interval (CI): 0.82-0.92] vs. 0.78 [95% CI: 0.73-0.84], P = 0.009). As the severity of calcification increased, both CCTA and FFR_CT showed reduced diagnostic efficacy, but FFR_CT maintained higher accuracy. In patients with greater than moderate calcification (AS ≥ 100), FFR_CT significantly outperformed CCTA (0.80 [95% CI: 0.74-0.85] vs. 0.62 [95% CI: 0.56-0.68], P < 0.001). Furthermore, integrating FFR_CT with CCTA and baseline factors using least absolute shrinkage and selection operator (LASSO) improved diagnostic performance in patients with greater than moderate calcification (AS≥100) than FFR_CT (0.85 [95% CI: 0.78-0.92] vs. 0.81 [95% CI: 0.72-0.90], P = 0.003).

Conclusion: FFR_CT offers superior diagnostic accuracy over CCTA, particularly in patients with moderate to severe calcification. Furthermore, the LASSO model enhances diagnostic performance in these cases, demonstrating potential for improving CAD diagnosis in patients with significant coronary calcification.