Clinical Study On Risk Stratification And Revascularization Of Stable Coronary Artery Disease Based On CT-FFR+FAI

Objective: To evaluate the clinical correlation between risk stratification and revascularization in patients with SCAD by CT-FFR+FAI technology integration.Method: Clinical data(including gender,age,systolic blood pressure,diastolic blood pressure,blood glucose,triglyceride,smoking history,drinking history and PCI history)and CCTA imaging data of 75 patients(225 vessels)with stable coronary heart disease diagnosed from July 2021 to December 2022 were retrospectively collected.At the same time,all CCTA image data were transmitted in DICOM format to AI analysis software(Shukun Science &Technology,China)for CT-FFR and FAI analysis.Kappa test was used to evaluate the consistency of subjective image quality scores and coronary artery stenosis rate grading results between two physicians.The variance analysis LSD method was used to analyze the difference of CT-FFR and FAI values among different risk levels.Independent sample t test was used to analyze the difference between CT-FFR and FAI values in patients with SCAD with or without revascularization.The predictive factors of stable coronary artery disease revascularization were analyzed by logistic regression,and the predictive efficacy of CT-FFR,FAI,combined CT-FFR and FAI for revascularization in stable coronary artery disease patients were analyzed by drawing the subjects’ working characteristic curves,respectively,and the optimal cut-off points of CT-FFR and FAI were obtained.Results: Among the 75 patients,excluding the subjective score of image quality ≤2 points(n=3),previous permanent pacemaker implantation(n=3),congenital malformation of coronary artery(n=5),and incomplete clinical data(n=4),60 patients were finally included.The results of subjective image quality score and coronary artery stenosis rate grading were consistent between the two physicians,with kappa values of 0.864 and 0.913,respectively.Subjective quality score: patient level(3 in 46 cases and 4 in 14 cases).Vascular level was 3in 121 and 4 in 59.Segwise,352 was 3 points and 188 was 4 points.Classification of coronary artery stenosis rate: 1 case was grade 0,5 cases were grade 1,16 cases were grade 2,13 cases were grade 3,20 cases were grade 4,and 5 cases were grade 5.Sub-group analysis of risk stratification showed that CT-FFR values of low,medium and high risk were ranked from high to low,and the difference was statistically significant.The FAI value in the low-risk group was lower in the higher-risk group,while the FAI value in the medium-risk group was lower in the higher-risk group,and the difference was statistically significant.Subgroup analysis of revascularization showed that CT-FFR value of revascularization group was lower than that of non-revascularization group,and FAI value of revascularization group was higher than that of non-revascularization group,and the difference was statistically significant.Logistic regression analysis showed that CT-FFR and FAI were independent predictors of revascularization in patients with SCAD.ROC curve showed that the area under the curve of CT-FFR was 0.773,and the optimal cut-off point was 0.710.The area under the curve of FAI was 0.753,the optimal cut-off point was-72 HU,and the area under the curve combined with CT-FFR and FAI was 0.827.Conclusion: The application of artificial intelligence CT-FFR and pericoronal FAI in risk stratification of patients with SCAD is feasible.When CT-FFR < 0.710 and pericoronal FAI >-72 HU,revascularization is recommended for SCAD patients,and the combination of CT-FFR and pericoronal FAI has significant incremental value in predicting whether SCAD patients will have revascularization.This indicates that the derived imaging technology can be used in clinical practice for a "one-stop" comprehensive and systematic evaluation of SCAD in anatomy,function and inflammation,so as to improve clinical suitability and economy.