Noninvasive Fractional Flow Reserve from Coronary CT Angiography

Transcription

1 2016 KSC Annual Spring Scientific Conference Noninvasive Fractional Flow Reserve from Coronary CT Angiography Bon-Kwon Koo, MD, PhD, Seoul, Korea

2 Why the hemodynamics for coronary artery disease? Twinlifemarketing.com.au

3 Q: Which is a significant stenosis? Please choose one. Coronary angiogram: left anterior descending coronary artery 3

4 This one should be a significant stenosis! Angiography: significant CT angiography: significant Intravascular ultrasound: significant Lumen area: 2.8mm 2 Vessel area: 9.0mm 2 Plaque burden: 69% Plaque Significant stenosis Ischemia/Chest pain Stent or surgery Better prognosis 4

5 Which is a significant stenosis? : Anatomy vs. Ischemia Stenosis severity by CT, angiography, intravascular US,... Extent of the perfusion territory Presence of myocardial infarction Myocardial blood flow including collaterals Microvascular function Physiologic or functional evaluation 5

6 Fractional Flow Reserve (FFR) Invasive physiologic test in a cath lab with very high spatial resolution Proximal Pressure (P a ) = 100mmHg Distal Pressure (P d ) = 70mmHg FFR=70/100=0.7 FFR 0.7 means that 30 % of myocardial blood flow was reduced due to the stenosis 6

7 Pressure Wire Hyperemic agent FFR vs. Myocardial ischemia R Not significant Significant stenosis

8 Q: Which is a significant stenosis? Please choose one. Coronary angiogram: left anterior descending coronary artery FFR = FFR = Anatomically insignificant, but hemodynamically significant! Anatomically significant, but hemodynamically insignificant! 8

9 FAME study PCI: percutaneous coronary intervention 2 Year Death/Myocardial infarction-free survival FFR (hemodynamics)-guided vs. Anatomy-guided - Less stent - Less cost - Same procedural time - Better clinical outcomes Tonino, et al. NEJM 2009; Pijls, et al. JACC

10 FFR is good for the patients and (relatively) simple. But, requires invasive procedure and expensive (>1,000 USD) cannot provide 3D anatomical information. LAD FFR 10

11 How to assess hemodynamics from static images? 3-D Model based on CCTA 11

12 Integration of non-invasive coronary imaging and hemodynamic lesion assessment 3-D Model based on CCTA Hybrid imaging: CCTA + SPECT/PET Stress CT perfusion imaging Transluminal attenuation gradient TAG = (HU/10mm) Estimation from geometry of stenosis 12

13 Computational Fluid Dynamics (CFD) Computational fluid dynamics (CFD) quantifies fluid pressure and velocity, based on physical laws of mass conservation and momentum balance. CFD is widely used in the aerospace and automotive industries for design and testing. Courtesy of C. Taylor, HeartFlow 13

14 CFD in simple and idealized coronary models Williams & Koo, et al. J Appl Physiol 2010 Na SH, Koo BK, et al. Korean Circ J

15 ccta + CFD = Pt-specific non-invasive FFR Feb 2010: Collaboration with CV SIM 2009: Project started 15

16 Patient-specific non-invasive FFR using CT & CFD Computational Model based on CCTA 3-D anatomic model from CCTA Blood Flow Solution Blood flow equations solved on supercomputer No additional imaging No additional medications Physiologic models -Myocardial demand -Morphometry-based boundary condition -Effect of adenosine on microcirculation Koo BK, EBC 2011, EuroPCR

17 Patient-specific non-invasive FFR using CT & CFD Computational Model based on CCTA 3-D anatomic model from CCTA Blood Flow Solution Blood flow equations solved on supercomputer No additional imaging No additional medications Physiologic models -Myocardial demand -Morphometry-based boundary condition -Effect of adenosine on microcirculation Koo BK, EBC 2011, EuroPCR

18 How can this novel technology change our daily practice? Current pathway CCTA Invasive angiography FFR FFR >50% diameter stenosis >50% diameter stenosis FFR 0.74 PCI FFR >50% diameter stenosis >50% diameter stenosis FFR 0.84 Medical treatment 18

19 How this novel technology can change our daily practice? Novel (risk-free, non-invasive, cost-saving) pathway CCTA FFR CT Invasive angiography and PCI 0.74 PCI >50% diameter stenosis FFR CT 0.74 Invasive procedures 0.85 Medical treatment FFR >50% diameter stenosis >50% diameter stenosis FFR 0.84 no ischemia 19

20 DISCOVER-FLOW study First-in-Human study To evaluate the feasibility and diagnostic performance of FFRCT Prospective, multicenter study 159 vessels in 103 patients Oct 2009 Jan 2011 Koo BK, et al, J Am Coll Cardiol,

21 DISCOVER-FLOW study Diagnostic performance of FFR CT and CCTA Per-vessel analysis (n=159) 25% PPV: positive predictive value, NPV: negative predictive value DISCOVER-FLOW: Koo BK, et al, J Am Coll Cardiol,

22 Clinical Evidences on Diagnostic Performance DISCOVER-FLOW 5 center FIH clinical trial Completed 2011 N=103 patients Published in JACC DeFACTO NXT 17 center clinical trial Completed 2012 N=252 patients Published in JAMA 10 center clinical trial Completed August, 2013 N=251 patients Published in JACC 22

23 Diagnostic performance of FFR CT Patient No Sensitivity Specificity PPV NPV Accuracy DISCOVER- FLOW % 82% 85% 91% 87% DeFACTO % 54% 67% 84% 73% NXT % 79% 65% 92% 81% Total: % 72% 72% 89% 80% 23

24 Specificity Non-invasive tests/ffr CT /Angiography vs. FFR 100% FFR CT -NXT 8 90% 80% MPI 2 70% Stress Echo 1 Angiography 7 FFR CT -DiscoverFLOW 6 Angiography 4 FFR CT -DeFACTO 4 60% Angiography 5 50% CCTA 4 40% CCTA 5 30% 30% 40% 50% 60% 70% 80% 90% 100% Sensitivity 1. Jung et al. EHJ 2008;29: Mellkian et al. JACC Intervention 2010;3: Muller et al. JACC Intervention 2011;4 4. Min et al. JAMA 2012;308: Meijboom et al. JACC 2008;52: Koo et al. JACC 2011;58: Park et al. JACC Norgaard et a. JACC

25 Clinical outcomes of FFRCT-guided decision Douglas PS, et al. EHJ

26 Clinical outcomes of FFRCT-guided decision Douglas PS, et al. EHJ

27 From CTA to CT-FFR and its beyond Planning the treatment strategy using Virtual revascularization & CT-derived computed FFR FFR CT after virtual stenting 27

28 After Stenting Before Stenting Planning the treatment strategy using Virtual revascularization & CT-derived computed FFR Angiography Invasive FFR CT-derived computed FFR (FFR CT ) FFR CT 0.12 FFR CT 0.11 FFR CT 0.72 FFR CT 0.02 Myocardial ischemia + Myocardial ischemia + Stent Kim KH, Koo BK, et al. JACC interv

29 After Stenting Before Stenting Planning the treatment strategy using Virtual revascularization & CT-derived computed FFR Angiography Invasive FFR CT-derived computed FFR (FFR CT ) FFR CT 0.12 FFR CT 0.11 FFR CT 0.72 FFR CT 0.02 Myocardial ischemia + Myocardial ischemia + Stent Stent FFR CT 0.86 No residual ischemia No residual ischemia Kim KH, Koo BK, et al. JACC interv

30 Diagnostic performance of FFRCT after virtual stenting to predict the residual ischemia Kim KH, Koo BK, et al. JACC interv

31 Virtual surgery before the surgery, with your computer FFR CT after one arterial graft and 2 saphenous vein grafts Koo BK, EuroPCR 31

32 Non-invasive hemodynamic measurement : Can we do more? Coronary CT angiography + Computational fluid dynamics Pressure Velocity Rest Hyperemia 32

33 Non-invasive hemodynamic force measurement : Is this feasible? Coronary CT angiography + Computational fluid dynamics Cauchy Stress Tensor Simulation Traction vector Wall Shear Stress (WSS) Velocity Pressure Oscillatory Shear Index (OSI) Particle Residence Time, Turbulent Kinetic Energy, 33

34 Hyperemic WSS [dyne/cm 2 ] FFR CT FFR 0.78 Hyperemic WSS [dyne/cm 2 ] FFR CT Hyperemic WSS Hyperemic WSS Non-invasive WSS assessment using ccta and CFD FFR CT FFR CT FFR Koo BK. International Symposium on Biomechanics

35 Association with Adverse plaque characteristics : WSS vs. % diameter stenosis Napkin ring sign Positive remodeling DS: % diameter stenosis, WSSR: resting wall shear stress, WSSH: hyperemic wall shear stress Park JB, Koo BK, Taylor C, et al. Submitted 35

36 Novel hemodynamic index: Axial Plaque Stress Axial plaque stress uniquely characterizes the diseased segment of both upstream and downstream. Choi GW Koo BK. JACC imaging

37 Stress [dyne/cm 2 ] Stress [dyne/cm 2 ] Distribution of Axial Plaque Stress in patients Upstream Upstream APS WSS Downstream Downstream FFR CT APS [dyne/cm 2 ] Radius Gradient APS (dyne/cm 2 ) APS WSS Upstream Upstream Downstream Downstream FFR CT APS [dyne/cm 2 ] Radius Gradient APS (dyne/cm 2 ) Choi GW Koo BK. JACC imaging

38 Influence of Lesion Shape on Hemodynamic Parameters (n=114) Upstream-dominant lesion (n=56) Downstream-dominant lesion (n=58) % Diameter Stenosis FFR CT Axial Plaque Stress (dyne/cm 2 ) 60% P<0.001 P< % 20% % Proximal segment Distal segment Choi GW Koo BK. JACC imaging

39 Frame Frame 106 Frame 101 Frame 96 Frame 96 Axial Plaque Stress (APS) Wall Shear Stress (WSS) Plaque rupture at segments 6 and 7 Choi GW, TPF workshop 2014, Incheon

40 CT, Asymptomatic APS and Future Event: CASE Acute MI Choi GW Koo BK. JACC imaging

41 APS and Future Event: CASE CT, Asymptomatic Acute MI APS Upstream 9960 dyne/cm 2 Downstream 1740 dyne/cm 2 Choi GW Koo BK. JACC imaging

42 Comprehensive non-invasive assessment for CAD using ccta and CFD Non-invasive hemodynamic assessment - Wall shear stress - Axial plaque stress APS [dyne/cm 2 ] Mauroich_Horvat P,, Hoffman U. Nat Rev Cardiol

43 Acknowledgement HeartFlow, USA: Charles Taylor, PhD, Gilwoo Choi, PhD, Hyun Jin Kim, PhD Seoul National University, Korea: Jun-Bean Park, MD, Do-Yeon Hwang, MD, Kyung-Jin Kim, MD Samsung Medical Center, Korea:, Joo-Myoung Lee, MD Inje university, Korea: Joon-Hyung Doh, MD, PhD Keimyung university, Korea: Chang-Wook Nam, MD, PhD Ulsan university, Korea: Eun-Seok Shin, MD, PhD Kobe university, Japan: Hiromasa Otake, MD, PhD