Clin Res Cardiol (2022). https://doi.org/10.1007/s00392-022-02002-5

Comparison of Computed Tomography Fractional Flow Reserve and conventional Anatomic Assessment in patients with severe aortic valve stenosis

H. S. Wienemann1, M. Langenbach2, V. Mauri1, M. Banazadeh1, S. Macherey-Meyer3, M. Meertens4, C. Hohmann1, S. Lee5, A. Hof6, E. Kuhn7, M. Halbach1, T. Wahlers7, S. Baldus4, A. Bunck2, M. Adam1

1Klinik III für Innere Medizin, Herzzentrum der Universität zu Köln, Köln; 2Radiologie, Universität zu Köln, Köln; 3Klinik III für Innere Medizin, Universität zu Köln, Medizinische Fakultät und Uniklinik, Köln; 4Klinik für Kardiologie, Angiologie, Pneumologie und Internistische Intensivmedizin, Herzzentrum der Universität zu Köln, Köln; 5Klinik III für Kardiologie, Angiologie, Pneumologie und Internistische Intensivmedizin, Universitätsklinikum Köln, Köln; 6Herzzentrum - Kardiologie, Universitätsklinikum Köln, Köln; 7Klinik und Poliklinik für Herz- und Thoraxchirurgie, Herzzentrum, Universitätsklinikum Köln, Köln;

Objective: Transcatheter aortic valve implantation (TAVI) is an established treatment method for patients with severe aortic stenosis (AS). In these patients, overall prevalence of coronary artery disease (CAD) is high. For TAVI, procedural planning computed tomography angiography (CT) is essential. These CT data may be used for anatomical assessment of coronary arteries and potential CAD evaluation. But anatomical evaluation alone might be challenging due to the lack of dedicated coronary CT sequences. Furthermore, differences between the anatomic CT-characterization of stenoses and their hemodynamic relevance might occur. For invasive coronary angiography, there is a large body of evidence that supports intracoronary physiology as the more accurate diagnostic method for PCI guiding than angiography alone. Hence, computed tomography derived fractional flow reserve (CT-FFR) might help to characterize lesion severity non-invasively from TAVI-CTs and reduce the referral rate to invasive coronary angiography (ICA). The aim of the study was to compare the diagnostic performance of TAVI-CT derived CT-FFR with CT-anatomical coronary assessment and the invasive non- hyperemic physiological index Resting Full-cycle Ratio (RFR) in patients with relevant AS and intermediate coronary stenoses. Materials and Methods: 22 patients with 24 coronary lesions underwent ICA with pressure wire assessment and routine pre-TAVI CT. Subsequently, we analyzed diagnostic quality of anatomical CT quantification as well as CT-FFR with respect to RFR as standard reference. CT-FFR analysis was performed using prototype on-site software (cFFR version 3.0, Siemens Healthineers). Results: The mean patient age was 81.3±5.4 years and 40.1 % were female. The mean aortic valve area was 0.75 ± 0.21 cm2, and the mean of peak aortic velocity was 4.2 ± 73.04 m/sec. The main artery interrogated was the left anterior descending artery, with 66.7% of the assessments, followed by the left circumflex artery with 25.0% and right coronary artery with 8.3%. The mean RFR was 0.90 ± 0.05. Positive RFR (≤0.89) was identified in 58.3% (n=14) of the lesions. On a per-lesion basis, diagnostic accuracy (DA), sensitivity (SE), specificity (SP), positive predictive value (PPV), and negative predictive value (NPV) of CT-FFR to discriminate lesion-specific ischemia defined by an RFR value of ≤0.89 were 70.8% (95% CI 48.9-87.4), 57.1% (95% CI 28.9-82.3), 90.0% (95% CI 55.5-99.7), 88.9% (95% CI 51.8- 99.7), and 60.0% (95% CI 32.3-83.7), respectively. In comparison, we calculated DA, SE, SP, PPV and NPV for CT-visual anatomical stenoses quantification >50%. (DA 58.3% (95% CI 36.6-77.9), SE 78.6% (95% CI 49.2-95.3), SP 30.0% (95% CI 6.7-65.2), PPV 6.11% (95% CI 35.7- 82.7), and NPV 50.0% (95% CI 18.8-88.2). The area under the receiver curve showed a larger area under the curve for CT-FFR (0.807; 95% CI 0.625–0.990) compared with CT stenoses of ≥50% (0.54; 95% CI 0.303–0.783) (Figure 1). Conclusion: Diagnostic planning CTs before TAVI can be used to apply anatomical graduation of coronary stenoses as well as physiological flow index calculations, even without dedicated cardiac coronary CT settings. Here, CT-FFR is superior to visual CT quantification for stenoses ≥50% in differentiating functional myocardial ischemia defined by invasive RFR in patients with AS. Figure 1:

https://dgk.org/kongress_programme/jt2022/aV1374.html