The impact of image quality on the diagnostic performance of CT-derived fractional flow reserve
10.3760/cma.j.cn112149-20211230-01165
- VernacularTitle:图像质量对冠状动脉CT血流储备分数诊断效能的影响
- Author:
Qingchao MENG
1
;
Yang GAO
;
Na ZHAO
;
Lei SONG
;
Hongjie HU
;
Tao JIANG
;
Wenqiang CHEN
;
Feng ZHANG
;
Lin LI
;
Li XU
;
Dumin LI
;
Lijuan FAN
;
Chaowei MU
;
Jingang CUI
;
Yunqiang AN
;
Bo XU
;
Bin LYU
Author Information
1. 中国医学科学院 北京协和医学院 国家心血管病中心 阜外医院放射影像科，北京 100037
- Keywords:
Tomography, X-ray computed;
Coronary stenosis;
Fractional flow reserve;
Image quality
- From:
Chinese Journal of Radiology
2023;57(2):150-156
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the impact of coronary CT angiography (CCTA) image quality and related factors on the diagnostic performance of CT-derived fractional flow reserve (CT-FFR).Methods:Based on the CT-FFR CHINA trial, the prospective multicenter trial enrolled patients with suspected coronary artery disease who underwent CCTA, CT-FFR and FFR measurement. The subjective and objective assessments of CCTA image were performed on a per-vessel level. The objective assessments included the enhancement degree of coronary artery, the signal-to-noise ratio (SNR) of the aortic root. We used χ 2 test and DeLong test to compare the diagnostic performance of CT-FFR with FFR as the reference standard in different subjective groups (non-artifact vs. artifact), enhancement degree of coronary artery groups (≤400 vs. 401-500 vs.>500 HU), SNR of the aortic root groups (≤16.9 vs.>16.9), body mass index (BMI) groups (<25 kg/m 2 vs.≥25 kg/m 2) and heart rate groups (<75 bpm vs.≥75 bpm). FFR and CT-FFR values≤0.80 was identified as myocardial ischemia. Results:The study enrolled 317 patients with 366 vessels. All target vessels in CCTA images were successfully analyzed by CT-FFR. The accuracy, sensitivity, specificity, positive predictive value, negative predictive value and AUC of the non-artifact group were 90.45%, 86.75%, 93.10%, 90.00%, 90.76% and 0.928, respectively, and those of the artifact group were 83.23%, 87.21%, 79.01%, 81.52%, 85.33% and 0.869, respectively. The differences in accuracy and specificity were statistically significant (χ 2=4.23, P=0.040; χ 2=8.55, P=0.003). The diagnostic efficacy of CT-FFR had no statistically significant differences among different objective groups (all P>0.05). Conclusions:The artifact of CCTA image has an effect on CT-FFR in the diagnosis of myocardial ischemia. The degree of vascular enhancement, SNR, BMI, and heart rate have no significant effect on the diagnostic performance of CT-FFR.
