Diagnostic accuracy of CT-derived and angiogram-derived fractional flow reserve

Although accumulating evidence demonstrated that virtual fractional flow reserve (FFR) based on coronary computed tomography angiography (CCTA) (CT-FFR) or invasive coronary angiogram (ICA) (CA-FFR) are promising alternatives to wire based FFR, which method has better diagnostic accuracy was still unclear. In our study, we aim to directly compare the diagnostic performance of CT-FFR and CA-FFR.

During the period of September 2019 to December 2020, patients with at least one 30%-90% coronary artery stenosis were enrolled and received invasive FFR. Then, virtual FFR values were calculated based on both CCTA and ICA, and then compared with the invasive FFR value.

Invasive FFR measurements were successfully performed in 114 vessels of 96 patients. Both CT-FFR and CA-FFR showed good correlation with wire-based FFR, with r values of 0.84 and 0.71 respectively. In paired t-test, the deviation of CT-FFR and CA-FFR was not significantly different (t = -1.9083, p = 0.05889). In Bland-Altman analysis, the coefficients of variation were 8.4% and 13.2% for CT-FFR and CA-FFR respectively. In ROC curve analysis, the per-vessel diagnostic accuracy of CT-FFR and CA-FFR was 94.7% and 92.1% respectively. The area under the curve of CT-FFR was slightly higher than that of CA-FFR (0.986 and 0.916 respectively, the difference between areas = 0.070, 95% CI 0.003-0.137, p = 0.0227).

Both CT-FFR and CA-FFR had good diagnostic concordance with wire-based FFR. In ROC Curve analysis, CT-FFR demonstrated slightly higher diagnostic accuracy than CA-FFR.

URL: https://www.chictr.org.cn/showproj.aspx?proj=44719. Unique Identifier: ChiCTR1900026971.