Impact of vascular morphology and plaque characteristics on computed tomography derived fractional flow reserve in early stage coronary artery disease

Background FFRCT gradually decreases from the proximal to the distal part of a vessel and reach the pathological threshold for significant ischemia even in the absence of obstructive coronary artery disease (CAD). The exact mechanisms of such gradual FFRCT decline remain unknown. Purpose The aims of this study are (1) to clarify the mechanisms of the gradual decline of computed tomography (CT) derived fractional flow reserve (FFRCT); and (2) to identify the predictive factors of an FFRCT decline below the pathological value of 0.80 in no apparent CAD vessels. Methods A total of 1058 outpatients with suspected CAD and who underwent CT angiography (CTA) with FFRCT analysis between January 2017 and December 2019 were evaluated. Among them, 150 consecutive patients who had both a CTA coupled to an FFRCT analysis and an invasive angiogram showing <25% coronary stenosis were included for analysis. Vessels were divided into two groups according to FFRCT at the distal vessel: FFRCT >0.80 (n=317) and FFRCT ≤0.80 (n=114). ΔFFRCT was defined as the magnitude of the change in FFRCT from the proximal to the distal vessel. Plaque characterization and vessel morphology measurements were performed semi-automatically. Vessel constituents were characterized based on Hounsfield units (HU) into lumen volume (<−50 HU), non-calcified plaque (NCP) (−50–150 HU), and calcified plaque (>150 HU). Results FFRCT decreased continuously from the proximal to distal across the three major vessels in both FFRCT>0.80 and FFRCT ≤0.80 groups (Figure 1). Compared to FFRCT>0.80 group, NCP volume was significantly higher in all three major vessels in FFRCT ≤0.80 group (210.2±83.6 mm3 vs. 140.9±139.3 mm3 for the RCA, p=0.01; 177.5±150.2 mm3 vs. 133.2±112.2 mm3 for the LAD, p=0.04; 127.6±91.5 mm3 vs. 58.7±57.7 mm3 for the LCX, p<0.01). Next, we investigated the vessel parameters that correlated with ΔFFRCT. ΔFFRCT was correlated with lumen volume in FFRCT>0.80 group (r=−0.24, p<0.0001), whereas ΔFFRCT was correlated with NCP volume in FFRCT ≤0.80 group (r=0.42, p<0.001) (Figure 2). An NCP volume above 44.8 mm3 was the strongest predictor of distal FFRCT of ≤0.80 (area under the curve 0.69, p<0.0001, sensitivity 95%, specificity 39%). Conclusions FFRCT is affected by vascular morphology and plaque characteristics even in the early stage of coronary artery disease. Our study highlights that subclinical coronary artery disease strongly influences FFRCT by effects unrelated to coronary stenosis. The presence of NCP is a major predictor of the gradual decrease of FFRCT toward pathological values. Anatomical findings as vessel morphology and plaque characteristics should be taken into consideration when interpreting numerical values of FFRCT to avoid unnecessary referrals for invasive coronary angiography or percutaneous coronary intervention. FUNDunding Acknowledgement Type of funding sources: None. Figure 1 Figure 2