Novel computational functional assessment of coronary stenosis and its clinical applications in predicting and evaluating procedural results
Ding, Daixin
Ding, Daixin
Loading...
Publication Date
2023-06-13
Type
Thesis
Downloads
Citation
Abstract
Coronary angiography has limited efficacy in identifying patients with suboptimal results after percutaneous coronary intervention (PCI). Immediate post-procedural functional assessment, including fractional flow reserve (FFR), is emerging as an effective tool for this purpose whereas with limited clinical adoption. Image-based computational FFR has been recently developed as an alternative without the need for costly pressure wire or hyperaemia-inducing medications. This thesis investigated the utility of image-based computational FFR assessment in predicting and evaluating the physiological efficacy of PCI. In the first part of the thesis, the relationship between post-PCI wire-based FFR and clinical outcomes was investigated using a systematic review and study-level meta analysis, pooling 12340 patients and 12923 vessels from 62 studies. Mean post-PCI FFR was not continuously associated with 1-year major adverse cardiac events (MACE) including all-cause death, myocardial infarction (MI), and target vessel revascularization (TVR). Still, risk ratio favoured high post-PCI FFR for reduced MACE, all-cause death, MI, TVR and better angina status under different cut-offs. The second part of the thesis focused on the utility of image-based FFR in post-PCI settings. The feasibility and accuracy of an optical coherence tomography (OCT)- based optical flow ratio (OFR) in predicting post-PCI FFR was evaluated. Post-PCI OFR was computed in 125 pullbacks from 119 patients with both OCT and FFR interrogation immediately after PCI. After eliminating the stenotic segment by virtual stenting, simulated residual OFR from pre-PCI OCT images was computed in 64 patients who had pre-PCI OCT. The accuracy in predicting post-PCI FFR ≤0.90 was 84% for post-PCI OFR and 80% for simulated residual OFR. Stent minimum expansion index was associated with in-stent pressure drop (r = -0.49, p<0.001). To overcome the restrictions of limited OCT pullback length, OFR was further incorporated with Murray law-based quantitative flow ratio (μQFR), a computational FFR from a single angiographic view, for a comprehensive physiological assessment of the entire interrogated vessel. The feasibility and accuracy of this OCT-modulated μQFR (OCT-μQFR) was validated in two datasets. Firstly, the diagnostic accuracy of single-view μQFR and two-view-based 3D-μQFR were compared in 280 vessels with two protocol-specified recommended angiographic views. μQFR computed from either angiographic view had similar diagnostic accuracy as compared with 3D-μQFR (92.1%, 92.5%, and 93.2%, respectively). μQFR from either angiographic view had excellent correlation (r = 0.95) and agreement (mean difference = 0.00 ± 0.03). Secondarily, the accuracy of OCT-μQFR as compared with μQFR was evaluated in 269 vessels with paired angiography, OCT, and FFR measurements. OCT-μQFR showed numerically higher diagnostic performance compared with μQFR (AUC = 0.95 versus 0.92, p = 0.057). The utility of OCT-μQFR for predicting physiological PCI efficacy was further investigated in 76 vessels from 74 patients undergoing OCT-guided PCI. Simulated residual OCT-μQFR was computed from pre-PCI co-registered angiography and OCT, by assuming full stent expansion to the intended-to-treat segment. Post-PCI OCT μQFR was computed from co-registered post-PCI angiography and OCT. Using actual post-PCI OCT-μQFR as reference, simulated residual OCT-μQFR showed good correlation (r = 0.80, p<0.001), agreement (mean difference = -0.02 ± 0.02, p<0.001) and diagnostic concordance (79%, using ≤0.90 for defining suboptimal functional stenting result). Post-PCI in-stent OCT-μQFR had a median value of 0.02 and was associated with LAD lesion location, higher baseline total plaque burden and fibrous plaque volume. To conclude, post-PCI wire-based FFR as a dichotomous variable was associated with clinical outcomes. Image-based computational FFR assessed immediately after PCI was feasible and accurate using FFR as reference. Estimating post-PCI physiology from baseline coronary images by virtual stenting was feasible and showed good concordance with post-PCI physiology. This thesis emphasized the potential application of image-based computational FFR in facilitating the achievement of optimal physiological efficacy of PCI.
Publisher
NUI Galway