Comparative Evaluation of Resting and Hyperemic Angiography‐Derived Index of Microcirculatory Resistance in Chronic Coronary Syndrome

Figures

• 

  Figure 1. Case example of angiography‐derived functional assessment. A, Selection of 2 series of images: The process begins with the acquisition of angiograms, which are transmitted in real‐time from the C‐arm machine to the XAscope software via the Picture Archiving and Communication System. Two sets of images focusing on the same target vessel are selected, ensuring they are captured from different angles (>25 degrees) to enhance the accuracy of subsequent analyses. The frames with the least noise are chosen for optimal target vessel segmentation. B, Three‐dimensional stereo reconstruction: Using a transformer‐based semantic segmentation network, this study innovatively achieves fully automated segmentation of the target vessel's main branch and its branches. A graph neural network is then used for automatic registration of the segmented vessels from the 2 angles. The data are transformed into a 3‐dimensional model through a binocular reconstruction algorithm, which facilitates a detailed structural analysis of the coronary anatomy. C, Hemodynamics simulation for xaFFR/xaIMR calculation: The flow rate in the target vessel is calculated by dividing the vessel's volume by the estimated…

• 

  Figure 2. Distribution of angiography‐derived IMR and wire‐derived IMR values. Violin plots illustrate the density trace of values for pressure wire–derived IMR, H‐xaIMR, and R‐xaIMR in the overall cohort of 139 vessels (A), subsequently segregated on the basis of CFR into CFR ≥2 (B) and CFR <2 (C). CFR indicates coronary flow reserve; H‐xaIMR, hyperemic x‐ray angiography–derived index of microcirculatory resistance; IMR, index of microcirculatory resistance; and R‐xaIMR, resting x‐ray angiography–derived index of microcirculatory resistance.

• 

  Figure 3. Correlation and agreement between angiography‐derived IMR and wire‐derived IMR. Scatterplot with regression line depicting the correlation between (A) H‐xaIMR and pressure wire–derived IMR and between (B) R‐xaIMR and pressure wire–derived IMR. Bland–Altman plot presenting the agreement between (C) H‐xaIMR and pressure wire–derived IMR and between (D) R‐xaIMR and pressure wire–derived IMR. H‐xaIMR indicates hyperemic x‐ray angiography–derived index of microcirculatory resistance; IMR, index of microcirculatory resistance; and R‐xaIMR, resting x‐ray angiography–derived index of microcirculatory resistance.

• 

  Figure 4. Comparison of different angiography‐derived IMR for the assessment of the coronary microcirculatory system. AUC demonstrates the efficiency of H‐xaIMR and R‐xaIMR in evaluating increased coronary microcirculatory resistance as determined by the IMR >25. H‐xaIMR exhibits significantly greater efficacy as compared with R‐xaIMR (P<0.001). AUC indicates area under the receiver operating characteristic curve; H‐xaIMR, hyperemic x‐ray angiography–derived index of microcirculatory resistance; IMR, index of microcirculatory resistance; and R‐xaIMR, resting x‐ray angiography–derived index of microcirculatory resistance.