Effects of stenotic lesion morphology on trans-lesional computed tomography derived fractional flow reserve (FFRCT) changes

Abstract

Abstract Background To assess the indication for revascularization CT-derived fractional flow reserve (FFRCT) has been measured at only a portion distal to the lesion. This conventional method is controversial in terms of its ability to accurately evaluate the coronary flow disturbance. Changes in FFRCT between the portions proximal and distal to the lesion (ΔFFRCT) are implemented as a novel method for the assessment of the coronary flow disturbance. The effect of different morphologies of stenotic lesions on ΔFFRCT remains unclear. Purpose To investigate the effect of lesion morphology on ΔFFRCT and the ability of ΔFFRCT to detect lesions with the coronary flow disturbance. Methods A total of 1502 outpatients with suspected stable coronary artery disease who underwent FFRCT between January 2017 and May 2022 were evaluated. Among them, 76 patients who underwent both FFRCT and invasive coronary angiography with ≥ 50% stenosis were evaluated and divided into four groups according to the degree of stenosis and lesion length (moderate, 50–69%; severe: 70–99%) and lesion length (focal, < 40 mm; diffuse, ≥ 40 mm) as follows: moderate/focal (n = 23), moderate/diffuse (n = 9), severe/focal (n = 34), and severe/diffuse (n = 10) (Figure 1). Vessel and lesion morphology (length, luminal volume, and plaque characteristics) and left ventricular mass were assessed. Indication for revascularization (percutaneous coronary intervention and coronary artery bypass graft) was assessed by invasive coronary angiography and additional invasive FFR measurement was performed when necessary. Results ΔFFRCT was greater in diffuse than in focal stenosis in both moderate and severe stenosis: moderate/focal (0.11 ± 0.12), moderate/diffuse (0.18 ± 0.10), severe/focal (0.26 ± 0.15), and severe/diffuse (0.38 ± 0.12) (Figure 2A). Lesion length correlated with ΔFFRCT (r = 0.31, p = 0.006) (Figure 2B). As for focal lesion, lesion length did not correlate with ΔFFRCT (r = 0.13, p = 0.34), whereas, as for diffuse lesion it did (r = 0.72, p = 0.0007) (Figure 2C). Multivariable analysis showed that lesion length was the strongest predictor of ΔFFRCT (β-coefficient = 0.41, p = 0.003), followed by lesion positive remodelling (β-coefficient = 0.27, p = 0.01), lesion calcified plaque volume (β-coefficient = －0.39, p = 0.02), and lesion spotty calcification (β-coefficient = 0.23, p = 0.04). ΔFFRCT predicted for lesion requiring revascularization (cut-off 0.21, AUC 0.80, sensitivity 64.4%, specificity 85.7%, 95% CI 0.69–0.91) (Figure 2D). Conclusions ΔFFRCT was affected by lesion length, followed by lesion calcium components. ΔFFRCT predicted lesions requiring revascularization.Figure 1Figure 2