Panagiotis K Siogkas1, Constantinos D Anagnostopoulos2, Riccardo Liga3,4, Themis P Exarchos5, Antonis I Sakellarios5, George Rigas5, Arthur J H A Scholte6, M I Papafaklis7, Dimitra Loggitsi8, Gualtiero Pelosi9, Oberdan Parodi9, Teemu Maaniitty10, Lampros K Michalis7, Juhani Knuuti10, Danilo Neglia9, Dimitrios I Fotiadis1,5,7. 1. Unit of Medical Technology and Intelligent Information Systems, Dept. of Materials Science and Engineering, University of Ioannina, Ioannina, Greece. 2. Center for Experimental Surgery, Clinical and Translational Research, Biomedical Research Foundation, Academy of Athens, 4 Soranou Ephessiou St., 115 27, Athens, Greece. cdanagnostopoulos@bioacademy.gr. 3. Cardio-Thoracic and Vascular Department, University Hospital of Pisa, Pisa, Italy. 4. Department of Nuclear Medicine, University Hospital Zurich, Zürich, Switzerland. 5. Biomedical Research Institute - FORTH, GR 45110 Ioannina, Ioannina, Greece. 6. Department of Cardiology, Leiden University Medical Center, Leiden, The Netherlands. 7. Michaelideion Cardiac Center, Dept. of Cardiology in Medical School, University of Ioannina, 451 10, Ioannina, Greece. 8. CT & MRI Department Hygeia-Mitera Hospitals, Athens, Greece. 9. Fondazione Toscana G. Monasterio and CNR Institute of Clinical Physiology, Pisa, Italy. 10. Turku PET Centre, University of Turku and Turku University Hospital, Turku, Finland.
Abstract
OBJECTIVES: Application of computational fluid dynamics (CFD) to three-dimensional CTCA datasets has been shown to provide accurate assessment of the hemodynamic significance of a coronary lesion. We aim to test the feasibility of calculating a novel CTCA-based virtual functional assessment index (vFAI) of coronary stenoses > 30% and ≤ 90% by using an automated in-house-developed software and to evaluate its efficacy as compared to the invasively measured fractional flow reserve (FFR). METHODS AND RESULTS: In 63 patients with chest pain symptoms and intermediate (20-90%) pre-test likelihood of coronary artery disease undergoing CTCA and invasive coronary angiography with FFR measurement, vFAI calculations were performed after 3D reconstruction of the coronary vessels and flow simulations using the finite element method. A total of 74 vessels were analyzed. Mean CTCA processing time was 25(± 10) min. There was a strong correlation between vFAI and FFR, (R = 0.93, p < 0.001) and a very good agreement between the two parameters by the Bland-Altman method of analysis. The mean difference of measurements from the two methods was 0.03 (SD = 0.033), indicating a small systematic overestimation of the FFR by vFAI. Using a receiver-operating characteristic curve analysis, the optimal vFAI cutoff value for identifying an FFR threshold of ≤ 0.8 was ≤ 0.82 (95% CI 0.81 to 0.88). CONCLUSIONS: vFAI can be effectively derived from the application of computational fluid dynamics to three-dimensional CTCA datasets. In patients with coronary stenosis severity > 30% and ≤ 90%, vFAI performs well against FFR and may efficiently distinguish between hemodynamically significant from non-significant lesions. KEY POINTS: Virtual functional assessment index (vFAI) can be effectively derived from 3D CTCA datasets. In patients with coronary stenoses severity > 30% and ≤ 90%, vFAI performs well against FFR. vFAI may efficiently distinguish between functionally significant from non-significant lesions.
OBJECTIVES: Application of computational fluid dynamics (CFD) to three-dimensional CTCA datasets has been shown to provide accurate assessment of the hemodynamic significance of a coronary lesion. We aim to test the feasibility of calculating a novel CTCA-based virtual functional assessment index (vFAI) of coronary stenoses > 30% and ≤ 90% by using an automated in-house-developed software and to evaluate its efficacy as compared to the invasively measured fractional flow reserve (FFR). METHODS AND RESULTS: In 63 patients with chest pain symptoms and intermediate (20-90%) pre-test likelihood of coronary artery disease undergoing CTCA and invasive coronary angiography with FFR measurement, vFAI calculations were performed after 3D reconstruction of the coronary vessels and flow simulations using the finite element method. A total of 74 vessels were analyzed. Mean CTCA processing time was 25(± 10) min. There was a strong correlation between vFAI and FFR, (R = 0.93, p < 0.001) and a very good agreement between the two parameters by the Bland-Altman method of analysis. The mean difference of measurements from the two methods was 0.03 (SD = 0.033), indicating a small systematic overestimation of the FFR by vFAI. Using a receiver-operating characteristic curve analysis, the optimal vFAI cutoff value for identifying an FFR threshold of ≤ 0.8 was ≤ 0.82 (95% CI 0.81 to 0.88). CONCLUSIONS: vFAI can be effectively derived from the application of computational fluid dynamics to three-dimensional CTCA datasets. In patients with coronary stenosis severity > 30% and ≤ 90%, vFAI performs well against FFR and may efficiently distinguish between hemodynamically significant from non-significant lesions. KEY POINTS: Virtual functional assessment index (vFAI) can be effectively derived from 3D CTCA datasets. In patients with coronary stenoses severity > 30% and ≤ 90%, vFAI performs well against FFR. vFAI may efficiently distinguish between functionally significant from non-significant lesions.