Comparative evaluation of the fractional flow reserve derived from coronary computed tomography angiography (CT-FFR) and fat attenuation index (FAI) in predicting revascularization

Abstract Purpose This study aimed to evaluate the clinical value of the fractional flow reserve derived from coronary computed tomography angiography (CT-FFR) and fat attenuation index (FAI) in predicting coronary revascularization. Methods Patients with known or suspected CAD who underwent coronary computed tomography angiography (CCTA) and subsequent invasive coronary angiography were screened. All CCTA data were calculated by a cloud workstation in standard Digital Imaging and Communications in Medicine format. Lesion-specific CT-FFR, distal-tip CT-FFR, and FAI were analyzed by core laboratories blinded to patient management. Results A total of 94 patients who received CCTA followed by invasive coronary angiography were identified and analyzed; 282 vessels were included for analysis. Overall, 54 (57.4%) patients with 72(25.5%) vessels demonstrated revascularization. In the multivariate model, FAI (odds ratio [OR]: 1.19; p < 0.001), lesion-specific CT-FFR (OR: 3.80; p = 0.009), and distal-tip CT-FFR (OR: 4.20; p = 0.008) values were identified as independent negative predictors. All receiver operating characteristic curves were above the reference line. The areas under the receiver operating characteristic curve for lesion-specific CT-FFR, distal-tip CT-FFR, and FAI were 0.798, 0.767, and 0.802, respectively. When the optimal threshold value of FAI was − 86 HU, the sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy for predicting revascularization were 88.9%, 59.0%, 42.7%, 93.2%, and 0.66, respectively. The corresponding values for the lesion-specific CT-FFR were 73.6%, 81.0%, 56.3%, 88.2%, and 0.78, respectively. Conclusions In patients with documented CAD on CCTA, adjunctive noninvasive functional testing based on the CT-FFR and FAI yielded similar overall accuracy for prediction of coronary revascularization. However, a significant difference was observed in diagnostic sensitivity of the FAI; the lesion-specific CT-FFR demonstrated the highest specificity. In conclusion, CT-FFR and FAI derived from quantitative CCTA improved the prediction of future revascularization. These parameters can potentially identify patients likely to require revascularization on referral for cardiac catheterization.