Part 1.Impact of Machine Learning Based Coronary Computed Tomography Angiography Derived Fractional Flow Reserve on Treatment Decision Making and Clinical Outcome in Patients with Suspected Coronary Disease Objective: To investigate the impact of coronary computed tomography angiography(CCTA)-derived machine learning(ML)-based fractional flow reserve(CT-FFR)compared to invasive coronary angiography(ICA)for therapeutic decision-making and clinical outcome in patients with suspected coronary artery disease.Methods:1121 consecutive patients with stable chest pain who underwent CCTA followed by ICA within 90 days over a ten-year period between January 2007 and December 2016 were included in this retrospective study.Patient follow-up for major adverse cardiac events(MACE)was performed.CT-FFR values were calculated using an artificial intelligence(AI)ML platform.Discordant results between severe stenosis via CT-FFR and severe stenosis on qualitative CCTA and ICA were also evaluated.The primary aim of the study was to assess the difference in trea CT-FFR,as a non-invasive diagnostic tool for assessing lesion-specific ischemia,directs therapeutic decision making with the potential to rationalize diagnostic workflows in patients with suspected CAD.tment strategy as directed by ICA alone versus CCTA with subsequent CT-FFR analysis.Results: CT-FFR was discordant with stenosis in 20.4%(229/1121)of patients for CCTA and 16.4%(184/1121)for ICA.After CT-FFR was revealed,ICA-based treatment regime change was seen in 167 patients(14.9%).Reserving ICA and revascularization for vessels with positive CT-FFR could reduce 54.5% ICA and lead to 4.4% fewer stent implantations.During median follow-up 26 months,positive CT-FFR was associated with MACE(HR:6.84,p<0.001),with superior prognostic value compared to severe stenosis on ICA(HR:1.84,p=0.002)and CCTA(HR:1.47,p=0.045).Conclusions: ML-based CT-FFR shows excellent performance in identifying patients with and without the need for invasive evaluation by ICA and better outcome prediction than severe anatomic stenosis on CCTA.CT-FFR,as a non-invasive diagnostic tool for assessing lesion-specific ischemia,directs therapeutic decision making with the potential to rationalize diagnostic workflows in patients with suspected CAD.Part 2.Prognostic Implication of CT-FFR Based Functional SYNTAX Score in Patients with Three-Vessel DiseaseObjective: The SYNTAX score(SS)is a pure anatomic score based on invasive coronary angiogram(ICA)and predicts outcome in patients with 3-vessel coronary artery disease(CAD).The feasibility and prognostic implication of “Functional SS(FSSCTA)” using a machine learning(ML)based CCTA derived fractional flow reserve(CT-FFR)in patients with three-vessel CAD have not yet been investigated.This study was aimed at investigating whether a ML based FSSCTA would predict clinical outcome in patients with three-vessel CAD.Methods: The SS based on CTA(SSCTA)and ICA(SSCTA)were retrospectively collected in 227 patients with three-vessel CAD.FSSCTA was derived from a ML based CT-FFR assessment.The ability of each score system to predict major adverse cardiac events(MACE)was compared.The difference between revascularization strategies directed by anatomical SS and FSSCTA was also assessed.Results:227 patients were divided into two groups according to the SSCTA cut-off value of 22.After determining FSSCTA for each patient,22.9% of patients(52/227)were reclassified to a low-risk group(FSSCTA?22).In the low vs.intermediate-to-high(>22)FSSCTA group,MACE occurred in 3.2%(4/125)vs.34.3%(35/102)respectively(p<0.001).The independent predictor of MACE was FSSCTA(OR=1.21,p=0.001)and diabetes(OR=2.3,p=0.048).FSSCTA demonstrated a better predictive accuracy for MACE compared with SSCTA(AUC:0.81 vs.0.75,p=0.01)and SSICA(0.81 vs.0.75,p<0.001).After FSSCTA was revealed,52 patients initially referred for CABG based on SSCTA would have been changed to PCI.Conclusions: Recalculating SS by incorporating lesion-specific ischemia as determined by ML based CT-FFR is a better predictor of MACE in patients with three-vessel CAD.Additionally,the use of FSSCTA may alter selected revascularization strategies in these patients.Part 3.Cinical Outcomes of Anatomical versus Functional Coronary CT Angiography for Coronary Artery DiseaseObjective: This study compared the downstream care and clinical outcomes of a machine learning(ML)-based fractional flow reserve derived from computed tomography(CT-FFR),“functional CCTA” compared with anatomical CCTA in patients with intermediate stenosis.Methods: Patients suspected coronary artery disease(CAD)with intermediate stenosis undergoing CCTA were included in this prospective study.Eligible patients were assigned to anatomical(CCTA)or functional CCTA(CT-FFR)group.The primary end point was the normalcy rate of invasive coronary angiography(ICA)within 90 days that showed non-obstructive stenosis(luminal stenosis?50%).Secondary end points are coronary revascularization and major adverse cardiovascular events(MACE).Results:533 subjects were allocated to the CT-FFR group and 532 to the anatomical CCTA group.At 90 days' follow-up,the rate of ICA demonstrating non-obstructive disease was higher in the anatomical CCTA group(28.8%,23/90)compared with that(2.1%,11/81)in the CT-FFR group(risk difference [RD]=15.2%,p=0.036).The referred ICA rate was significantly higher in the CCTA group(24.2%)than CT-FFR group(18.8%)(RD=5.4%,p=0.029).The revascularization(REV)-to-ICA ratio was30.2%(39/90)for those in the CCTA group,which was lower than the CT-FFR group(63.0%,51/81)(RD=19.7%,p<0.001).MACE was significantly more common among patients in the CCTA group compared with patients in the CT-FFR group(HR:1.76;95% CI: 1.02 to 3.31,p=0.046).Conclusions: In stable patients with intermediate stenosis,the CT-FFR strategy was associated with a lower rate referral for ICA,a normalcy rate of ICA demonstrating non-obstructive disease,and better clinical outcome than the anatomical CCTA strategy.This supports the usefulness of functional CCTA as an efficient method to guide patient management. |