Yin Zhuhao, Zhou Changsheng, Guo Jian, Wei Yuan, Ma Yifei, Zhou Fan, Zhu Wusheng, Zhang Long Jiang
Department of Radiology, Jinling Hospital, Nanjing Medical University, Nanjing, Jiangsu 210002, China.
Department of Radiology, Jinling Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, Jiangsu 210002, China.
Eur J Radiol. 2024 Feb;171:111285. doi: 10.1016/j.ejrad.2024.111285. Epub 2024 Jan 3.
CT-derived fractional flow reserve (CT-FFR) has been widely applied in coronary hemodynamic assessment. However, the feasieablity and standardization measurement in intracranial artery stenosis (ICAS) remains to be defined.
To demonstrate the feasibility of CT-FFR in ICAS functional assessment and explore the optimal CT-FFR measurement position with invasive FFR as reference standard.
Nineteen patients (mean age, 58.6 years ± 1.9 [SD]; 13 men) with moderate to severe (≥50 %) ICAS undergoing guidewire-based pressure measurement and preoperative head CT angiography (CTA) were retrospectively enrolled. CT-FFR was measured in the following standard measurement positions, including the end of stenosis (D0), 1 cm distal to the stenosis (D1) and 2 cm distal to the stenosis (D2). Diagnostic performance of CT-FFR was assessed by the area under the curve (AUC) of receiver operating characteristic curves by assuming invasive FFR ≤ 0.80 or 0.75 as hemodynamically significant stenosis.
Excellent intra- and inter-observer agreement (ICC range, 0.930-0.992) was observed for CT-FFR measurement in different positions. Under different FFR thresholds, the diagnostic performance of CT-FFR showed perfect prediction with AUC values of 1.000 (95 % CI: 0.824, 1.000). The sensitivity, specificity and AUC of CT-FFR ≤ 0.80 in detecting FFR ≤ 0.80 was 0.94 (95 % CI: 0.68, 1.00), 1.00 (95 % CI: 0.31, 1.00) and 0.969 (95 % CI: 0.772, 1.000), respectively. Similar performance of CT-FFR ≤ 0.75 was obtained for identifying FFR ≤ 0.75 with the AUC of 0.964. The strongest correlation (r = 0.915, p < 0.001) and agreement (mean difference: 0.02, 95 % limits of agreement: -0.16 to 0.19) were observed between CT-FFR and FFR.
Cerebral CT-derived fractional flow reserve (CT-FFR) measured 1 cm distal to stenosis achieved the most comparable results with invasive FFR, which indicated its potentially promising clinical application for evaluating the functional relevance of intracranial artery stenosis.
基于CT的血流储备分数(CT-FFR)已广泛应用于冠状动脉血流动力学评估。然而,颅内动脉狭窄(ICAS)的可行性及标准化测量仍有待明确。
以有创FFR作为参考标准,论证CT-FFR在ICAS功能评估中的可行性,并探索最佳CT-FFR测量位置。
回顾性纳入19例中度至重度(≥50%)ICAS患者(平均年龄58.6岁±1.9[标准差];13例男性),这些患者均接受了基于导丝的压力测量及术前头颅CT血管造影(CTA)。在以下标准测量位置测量CT-FFR,包括狭窄末端(D0)、狭窄远端1 cm处(D1)和狭窄远端2 cm处(D2)。通过假设侵入性FFR≤0.80或0.75为血流动力学显著狭窄,采用受试者操作特征曲线的曲线下面积(AUC)评估CT-FFR的诊断性能。
不同位置CT-FFR测量的观察者内及观察者间一致性良好(ICC范围为0.930 - 0.992)。在不同的FFR阈值下,CT-FFR的诊断性能显示出完美预测,AUC值为1.000(95%CI:0.824,1.000)。CT-FFR≤0.80检测FFR≤0.80的敏感性、特异性和AUC分别为0.94(95%CI:0.68,1.00)、1.00(95%CI:0.31,1.00)和0.969(95%CI:0.772,1.000)。对于识别FFR≤0.75,CT-FFR≤0.75的性能相似,AUC为0.964。CT-FFR与FFR之间观察到最强相关性(r = 0.915,p < 0.001)和一致性(平均差异:0.02,95%一致性界限:-0.16至0.19)。
在狭窄远端1 cm处测量的脑CT衍生血流储备分数(CT-FFR)与侵入性FFR的结果最具可比性,这表明其在评估颅内动脉狭窄功能相关性方面具有潜在的临床应用前景。