National Heart Centre Singapore, Singapore.
Duke-NUS Medical School, National University of Singapore, Singapore.
Am J Physiol Heart Circ Physiol. 2020 Aug 1;319(2):H360-H369. doi: 10.1152/ajpheart.00135.2020. Epub 2020 Jul 17.
Proper inlet boundary conditions are essential for accurate computational fluid dynamics (CFD) modeling. We developed methodology to derive noninvasive FFR using CFD and computed tomography coronary angiography (CTCA) images. This study aims to assess the influence of brachial mean blood pressure (MBP) and total coronary inflow on FFR computation. Twenty-two patients underwent both CTCA and FFR measurements. Total coronary flow was computed from left ventricular mass (LVM) measured from CTCA. A total of 286 CFD simulations were run by varying MBP and LVM at 70, 80, 90, 100, 110, 120, and 130% of the measured values. FFR increased with incrementally higher input values of MBP: 0.78 ± 0.12, 0.80 ± 0.11, 0.82 ± 0.10, 0.84 ± 0.09, 0.85 ± 0.08, 0.86 ± 0.08, and 0.87 ± 0.07, respectively. Conversely, FFR decreased with incrementally higher inputs value of LVM: 0.86 ± 0.08, 0.85 ± 0.08, 0.84 ± 0.09, 0.84 ± 0.09, 0.83 ± 0.10, 0.83 ± 0.10, and 0.82 ± 0.10, respectively. Noninvasive FFR calculated using measured MBP and LVM on a total of 30 vessels was 0.84 ± 0.09 and correlated well with invasive FFR (0.83 ± 0.09) ( = 0.92, < 0.001). Positive association was observed between FFR and MBP input values (mmHg) and negative association between FFR and LVM values (g). Respective slopes were 0.0016 and -0.005, respectively, suggesting potential application of FFR in a clinical setting. Inaccurate MBP and LVM inputs differing from patient-specific values could result in misclassification of borderline ischemic lesions. While brachial mean blood pressure (MBP) and left ventricular mass (LVM) measured from CTCA are the two CFD simulation input parameters, their effects on noninvasive fractional flow reserve (FFR) have not been systematically investigated. We demonstrate that inaccurate MBP and LVM inputs differing from patient-specific values could result in misclassification of borderline ischemic lesions. This is important in the clinical application of noninvasive FFR in coronary artery disease diagnosis.
准确的入口边界条件对于计算流体动力学（CFD）建模至关重要。我们开发了一种使用 CFD 和冠状动脉计算机断层扫描血管造影（CTCA）图像无创计算 FFR 的方法。本研究旨在评估肱动脉平均血压（MBP）和总冠状动脉血流量对 FFR 计算的影响。22 名患者均同时接受了 CTCA 和 FFR 测量。总冠状动脉流量是根据 CTCA 测量的左心室质量（LVM）计算得出的。通过将 MBP 和 LVM 分别测量值的 70%、80%、90%、100%、110%、120%和 130%作为输入值，共进行了 286 次 CFD 模拟。FFR 随着输入 MBP 的增加而增加：0.78±0.12、0.80±0.11、0.82±0.10、0.84±0.09、0.85±0.08、0.86±0.08 和 0.87±0.07。相反，FFR 随着输入 LVM 的增加而降低：0.86±0.08、0.85±0.08、0.84±0.09、0.84±0.09、0.83±0.10、0.83±0.10 和 0.82±0.10。在总共 30 条血管上使用测量的 MBP 和 LVM 计算得出的无创 FFR 为 0.84±0.09，与有创 FFR（0.83±0.09）相关性良好（r=0.92，p<0.001）。FFR 与 MBP 输入值（mmHg）呈正相关，与 LVM 值（g）呈负相关。各自的斜率分别为 0.0016 和-0.005，这表明 FFR 在临床应用中的潜在应用。与患者特定值不同的不准确的 MBP 和 LVM 输入可能导致边界缺血性病变的错误分类。虽然从 CTCA 测量的肱动脉平均血压（MBP）和左心室质量（LVM）是两个 CFD 模拟输入参数，但它们对无创性分流量储备（FFR）的影响尚未得到系统研究。我们证明，与患者特定值不同的不准确的 MBP 和 LVM 输入可能导致边界缺血性病变的错误分类。这在冠心病诊断中无创 FFR 的临床应用中很重要。
