Construction And Clinical Application Of CT-FFR Based Functional Duke Jeopardy Score

Part 1 Construction and validation of CT-FFR based functional Duke Jeopardy Score systemBackgroundCT derived fractional flow reserve(CT-FFR)has emerged as a non-invasive diagnostic tool that can evaluate hemodynamically significant coronary artery stenosis,guide treatment strategies,and provide prognostic information.However,a substantial proportion of patients with pathologic CT-FFR results do not experience major adverse cardiovascular events(MACE)during short-and mid-term follow-up,indicating that CT-FFR results should be further refined.The feasibility of calculating Duke Jeopardy Score(DJS)by Invasive coronary angiography(ICA)and coronary CT angiography(CCTA)has been demonstrated.However,the role of integrating CT-FFR with anatomical DJS to predict prognostic outcomes has yet to be explored.This study aims to explore the reproducibility of CT-FFR based functional DJS(fDJSCTA)and evaluate its potential in predicting MACE in suspected coronary artery disease(CAD)patients.MethodsThe study retrospectively included 252 patients who underwent who underwent invasive flow reserve fraction(FFR)within 90 days after coronary CT angiography(CCTA).DJS was used to evaluate the amount of ischemic myocardium based on the invasive FFR(fDJSICA),CT-FFR(fDJSCTA),and CCTA(DJSCTA)results.The Spearman correlation and Bland-Altman analysis were used to analyze the correlation between fDJSCTA and fDJSICA.The interobserver and intraobserver agreement of the fDJSCTA was assessed using the intraclass correlation coefficient(ICC).The optimal cutoff values of DJSCTA and fDJSCTA for predicting MACE were evaluated by using maximally selected rank statistics methods.Prognostic analysis was performed using uni variable and multivariable Cox proportional hazards regressions.Hazard Ratio(HR),95%Confidence interval(CI)and Concordance Index(C-index)were used to represent the performance of the prediction model.ResultsThere was a high correlation between fDJSCTA and fDJSICA(Spearman r=0.82;95%CI:0.77-0.86).Bland-Altman analysis showed a slightly systematic overestimation of DJS values with fDJSCTA(mean difference:0.294;95%limits of agreement:0.128-0.459).There was good interobserver and intraobserver reproducibility of fDJSCTA(ICCs:0.79[95%CI:0.63-0.88;p<0.001],0.85[95%CI:0.74-0.92;p<0.001],respectively).Using the maximally selected rank statistics method,the optimal cutoff values of DJSCTA and fDJSCTA to predict the occurrence of MACE were>0 and≥ 4,respectively.Thus,252 patients were divided into two groups according to the optimal cutoff value of fDJSCTA:low fDJSCTA group and high fDJSCTA group.The rates of MACE were significantly higher in the high fDJSCTA group(30.6%,22/72)than in the low fDJSCTA group(3.9%,7/180)(p<0.001).A multivariate Cox model for MACE in function of the fDJSCTA cutoff resulted in an adjusted HR of 9.05(95%CI:3.73-23.71,p<0.001).The prediction model showed a C-index of 0.784(95%CI:0.727-0.834)to predict MACE.Conclusions There was a good correlation between the fDJSCTA and fDJSICA.There was good interobserver and intraobserver reproducibility of fDJSCTA.fDJSCTA is an independent predictor of MACE in patients with CAD.Part 2 Validation of the CT-FFR based functional Duke Jeopardy Score system for clinical decision making and prognosis assessment in patients with suspected coronary artery diseaseBackgroundIn our previous study,it was found that Functional Duke Jeopardy Score(fDJSCTA)based on CT-FFR was associated with adverse clinical outcomes in patients with CAD.However,further validation of the prognostic value of fDJSCTA for adverse clinical outcomes in patients with CAD is still needed.Furthermore,the value of fDJScTA in improving clinical decision-making and reducing unnecessary invasive testing in CAD patients remains unclear.Herein,the aim of this study was to validate a novel CT-FFR based functional DJS(fDJSCTA)and evaluate the potential of fDJSCTA in predicting adverse outcomes and guiding downstream management decisions in CAD patients.MethodsThis study was a secondary post hoc analysis of a prospective single-center,two-arm controlled study and finally included 566 participants with 25%-80%stenosis of at least one main coronary artery in CCTA.In this study,follow-up after enrollment was obtained by medical record review,clinic visits,and telephone contact at 90 days,1 year and 3 years,respectively.Based on the fDJSCTA cutoff value in section 1,Kaplan-Meier analysis was used to estimate the cumulative incidence of clinical events,and log-rank testing was used for comparison.Univariate and multivariate Cox proportional hazards survival models reporting a hazard ratio(HR)with 95%confidence interval(CI)were calculated.We also assessed the ability of each of the methods(CT-FFR,DJSCTA,fDJSCTA as continuous variables and CCTA stenosis as three categories:20%-49%,50-69%and≥ 70%)to predict outcomes by using time-dependent area under the receiver operating characteristics curve(AUC)statistics.The risk reclassification of CT-FFR by fDJSCTA was assessed by continuous net reclassification improvement(NRI).To explore the differences between CT-FFR and fDJSCTA based revascularization strategies.ResultsIn this section,fDJSCTA was the strongest independent predictor for MACE(adjusted HR:7.08[95%CI:2.96-16.93];P<0.001)during a median 3-year follow-up.Kaplan-Meier survival curves of all patients,patients with CT-FFR≤0.80 and patients with CT-FFR≤0.70 indicated that the prognosis was significantly better in the low-risk group than in the high-risk group(both P<0.05).fDJSCTA performed well at different follow-up time points with the AUCs of 0.89,0.83,0.82 at 360,720,and 1080 days,respectively,and continuously superior to other methods.fDJSCTA resulted in risk stratification changes from the per-patient CT-FFR model in 19.7%(88/447)of patients,leading to more accurate MACE prediction when compared with MACE occurrences in the real-world setting(NRI=0.155;P<0.001).Based on the treatment strategy directed by fDJSCTA,the number of ICA referrals dropped by 47.4%,and the rate of revascularization could have been increased by 12.6%.ConclusionIn summary,this study validated a CT-FFR based functional Duke jeopardy scoring system for assessing ischemic myocardial burden and to evaluate its value in prognosis and treatment decisions.Compared with CT-FFR alone,the CT-FFR based functional Duke jeopardy scoring system provides a novel,simple,effective and personalized method to stratify the risk of MACE and assist therapeutic decision-making.