Wang Wei, Hu Yumeng, Xiang Peng, Sheng Xiaosheng, Leng Xiaochang, Yang Xinyi, Dong Liang, Li Changling, Sun Yong, Jiang Jun, Xiang Jianping, Lin Xianhe
Department of Cardiology, The First Affiliated Hospital of Anhui Medical University, Hefei, China.
ArteryFlow Technology Co., Ltd., Hangzhou, China.
Quant Imaging Med Surg. 2023 Apr 1;13(4):2496-2506. doi: 10.21037/qims-22-893. Epub 2023 Mar 6.
A computational method (AccuFFrangio) based on invasive coronary angiography (ICA) and computational fluid dynamics (CFD) to calculate fractional flow reserve (FFR) without a pressure wire has been devised to clarify the physiological significance of coronary stenosis. This study aimed to evaluate the diagnostic performance of AccuFFRangio computation under different boundary conditions and vessel reconstruction approaches.
Consecutive patients with stable angina pectoris who underwent ICA and FFR assessment from 2 centers were analyzed retrospectively. Using wire-based FFR as the reference standard, the diagnostic performances of AccuFFRangio and its variations were evaluated and compared. The calculation of AccuFFRangio involves several key boundary conditions, including patient-specific aortic pressure, contrast flow velocity derived from the thrombolysis in myocardial infarction (TIMI) frame count method, and vessel reconstruction based on 2 angiographic views. We considered the following 3 variations: (I) a fixed aortic pressure [fixed pressure AccuFFRangio (pAccuFFRangio)], (II) an empirical hyperemic velocity [fixed velocity AccuFFRangio (vAccuFFRangio)], and (III) vessel reconstruction using a single angiographic view [single view AccuFFRangio (sAccuFFRangio)].
A total of 230 patients with 230 vessels were included in the final analysis. The accuracy for standard AccuFFRangio, pAccuFFRangio, vAccuFFRangio, and sAccuFFRangio was 93.91%, 86.52%, 81.74%, and 83.48%, respectively; the sensitivity was 90.74%, 51.85%, 83.33%, and 46.30%, respectively; the specificity was 94.89%, 97.16%, 81.25%, and 94.89%, respectively; and the area under the receiver operating characteristic curve was 0.971, 0.928, 0.892, and 0.870, respectively.
The comparison suggested that the overall performance of the standard AccuFFRangio was superior to other variations and had the highest accuracy among all the cases.
已设计出一种基于有创冠状动脉造影（ICA）和计算流体动力学（CFD）的计算方法（AccuFFrangio），用于在不使用压力导丝的情况下计算血流储备分数（FFR），以阐明冠状动脉狭窄的生理意义。本研究旨在评估不同边界条件和血管重建方法下AccuFFRangio计算的诊断性能。
回顾性分析来自2个中心的连续接受ICA和FFR评估的稳定型心绞痛患者。以基于导丝的FFR作为参考标准，评估并比较AccuFFRangio及其变体的诊断性能。AccuFFRangio的计算涉及几个关键边界条件，包括患者特异性主动脉压力、源自心肌梗死溶栓（TIMI）帧数法的造影剂流速，以及基于2个血管造影视图的血管重建。我们考虑了以下3种变体：（I）固定主动脉压力[固定压力AccuFFRangio（pAccuFFRangio）]，（II）经验性充血速度[固定速度AccuFFRangio（vAccuFFRangio）]，以及（III）使用单个血管造影视图进行血管重建[单视图AccuFFRangio（sAccuFFRangio）]。
最终分析共纳入230例患者的230条血管。标准AccuFFRangio、pAccuFFRangio、vAccuFFRangio和sAccuFFRangio的准确率分别为93.91%、86.52%、81.74%和83.48%；敏感性分别为90.74%、51.85%、83.33%和46.30%；特异性分别为94.89%、97.16%、81.25%和94.89%；受试者工作特征曲线下面积分别为0.971、0.928、0.892和0.870。
比较表明，标准AccuFFRangio的总体性能优于其他变体，在所有情况中准确率最高。
