Clinical Validation of a Virtual Planner for Coronary Interventions Based on Coronary CT Angiography
Jeroen Sonck, Sakura Nagumo, Bjarne Linde Nørgaard, Hiromasa Otake, Brian Ko, Jinlong Zhang, Takuya Mizukami, Michael Mæng, Daniele Andreini, Yu Takahashi, Jesper Møller Jensen, Abdul Rahman Ihdayhid, Ward Heggermont, Emanuele Barbato, Niya Mileva, Daniel Munhoz, Jozef Bartúnek, Adam Updegrove, Amy Collinsworth, Martin Pěnička, Lieven Van Hoe, Jonathon Leipsic, Bon‐Kwon Koo, Bernard De Bruyne, Carlos Collet
Abstract
Low fractional flow reserve (FFR) values after percutaneous coronary intervention (PCI) carry a worse prognosis than high post-PCI FFR values. Therefore, the ability to predict post-PCI FFR might play an important role in procedural planning. Post-PCI FFR values can now be computed from pre-PCI coronary computed tomography angiography (CTA) using the fractional flow reserve derived from coronary computed tomography angiography revascularization planner (FFRCT Planner). The aim of this study was to validate the accuracy of the FFRCT Planner. In this multicenter, investigator-initiated, prospective study, patients with chronic coronary syndromes and significant lesions based on invasive FFR ≤0.80 were recruited. The FFRCT Planner was applied to the fractional flow reserve derived from coronary computed tomography angiography (FFRCT) model, simulating PCI. The primary objective was the agreement between the predicted post-PCI FFR by the FFRCT Planner and measured post-PCI FFR. Accuracy of the FFRCT Planner’s luminal dimensions was assessed by using post-PCI optical coherence tomography as the reference. Overall, 259 patients were screened, with 120 patients (123 vessels) included in the final analysis. The mean patient age was 64 ± 9 years, and 24% had diabetes. Measured FFR post-PCI was 0.88 ± 0.06, and the FFRCT Planner FFR was 0.86 ± 0.06 (mean difference: 0.02 ± 0.07 FFR unit; limits of agreement: –0.12 to 0.15). Optical coherence tomography minimal stent area was 5.60 ± 2.01 mm2, and FFRCT Planner minimal stent area was 5.0 ± 2.2 mm2 (mean difference: 0.66 ± 1.21 mm2; limits of agreement: –1.7 to 3.0). The accuracy and precision of the FFRCT Planner remained high in cases with focal and diffuse disease and with low and high calcium burden. The FFRCT-based technology was accurate and precise for predicting FFR after PCI. (Precise Percutaneous Coronary Intervention Plan Study [P3]; NCT03782688)