Diagnostic performance of non-invasive fractional flow reserve derived from coronary CT angiography in suspected coronary artery disease: The NXT trial Bjarne L. Nørgaard, Jonathon Leipsic, Sara Gaur, Sujith Seneviratne, Brian S. Ko, Hiroshi Ito, Jesper M. Jensen, Laura Mauri, Bernard De Bruyne, Hiram Bezerra, Kazuhiro Osawa, Mohamed Marwan, Christoph Naber, Andrejs Erglis, Seung-Jung Park, Evald H. Christiansen, Anne Kaltoft, Jens F. Lassen, Hans Erik Bøtker, Stephan Achenbach For the NXT Investigators
Disclosures Study funding provided by HeartFlow which had no involvement in the data analysis, abstract planning or preparation No study investigator had any financial interest related to the study sponsor
Background Coronary CT Angiography: High diagnostic accuracy for anatomic stenosis detection Cannot determine physiologic significance of lesions 1 1 Min et al. J Am Coll Cardiol 2010; 55: 957
Background Coronary CT Angiography: High diagnostic accuracy for anatomic stenosis detection Cannot determine physiologic significance of lesions 1 Invasive Fractional Flow Reserve (FFR): Gold standard for diagnosis of lesion-specific ischemia 2 Improves event-free survival and cost effectiveness 3,4 1 Min et al. J Am Coll Cardiol 2010; 55: 957; 2 Piljs et al. Cath Cardiovasc Interv 2000;49:1; 3 Tonino et al. N Engl J Med 2009;360:213; 4 Berger et al. J Am Coll Cardiol 2005;46:438
Background Coronary CT Angiography: High diagnostic accuracy for anatomic stenosis detection Cannot determine physiologic significance of lesions 1 Invasive Fractional Flow Reserve (FFR): Gold standard for diagnosis of lesion-specific ischemia 2 Improves event-free survival and cost effectiveness 3,4 FFR computed from standard acquired coronary CT angiography images (FFR CT ) 5,6 1 Min et al. J Am Coll Cardiol 2010; 55: 957; 2 Piljs et al. Cath Cardiovasc Interv 2000;49:1; 3 Tonino et al. N Engl J Med 2009;360:213; 4 Berger et al. J Am Coll Cardiol 2005;46:438; 5 Koo BK et al. J Am Coll Cardiol; 2011;58:1989; 6 Min J et al. JAMA 2012;308:237.
FFR CT Technology Patient-Specific Coronary Flow and Pressure: Using a standard CT dataset a quantitative model is built A physiological model is developed using LV and coronary anatomy and established form-function principles A fluid model calculates flow and pressure under simulated hyperemic conditions Taylor et al, JACC 2013; 61: 2233-41
Study Objectives To determine the diagnostic performance of non-invasive FFR CT using invasive FFR as the reference standard To compare the diagnostic performance of FFR CT vs. anatomic testing (coronary CTA or invasive coronary angiography) Incorporates learnings from previous FFR CT trials: Newest generation of FFR CT analysis software CT acquisition according to societal guidelines 1 1 Abbara S et al. J Cardiovasc Comput Tomography 2009;3:190;
Study Endpoints Primary Endpoint: Per-patient diagnostic performance as assessed by the area under the receiver operating characteristic curve (AUC) of FFR CT vs. coronary CTA for the diagnosis of ischemia. (Reference standard: FFR 0.80) Secondary Endpoints: Diagnostic performance (accuracy, sensitivity, specificity, PPV and NPV) of FFR CT, coronary CTA, and invasive coronary angiography
Subject Inclusion / Exclusion Criteria Inclusion Criteria: Underwent >64-row CT and ICA scheduled < 60 days between CT and ICA Exclusion Criteria: Prior CABG or PCI Suspected ACS Recent MI within 30 days of CT Contraindication to nitrates, beta blockade or adenosine ICA = Invasive coronary angiography; CABG = coronary artery bypass surgery; ACS = acute coronary syndrome; MI = myocardial infarction; PCI = percutaneous coronary intervention
Study Procedures Blinded core laboratories for FFR and FFR CT CT: Acquisition protocols according to societal guidelines 1 Image quality independently evaluated via predefined scoring system 2 Positive: Site-read stenosis severity >50% 3 ICA: Positive: Site-read stenosis severity >50% FFR: At maximum hyperemia during ICA Adenosine 140 180 mcg/kg/min IV Positive: 0.80 4 1 Abbara S et al. J Cardiovasc Comput Tomography 2009;3:190; 2 Gaur S et al Cardiovasc Comput Tomogr 2013, in press; 3 Raff GL et al. J Cardiovasc Comp Tomogr 2009; 3: 122-36; 4 Tonino PA et al. N Engl J Med 2009; 360: 213-24;
Patient Enrollment Study enrollment 9/2012 8/2013 10 sites in Europe, Asia, and Australia Screened Cohort (n=357) Secondary endpoints Patients n=254 Vessels n=484 Excluded Image artifacts (n=44, 13%) Failed inclusion criteria (n=27) FFR quality (n=22) Other (n=10) Excluded from Primary Endpoint No ccta stenosis 30-90% (n=3) Primary Endpoint Patients n=251
Study Population CT Characteristics Patient Characteristics Age (years) [mean + SD] 64 ± 10 Male gender 64% Prior MI 2% Diabetes mellitus 23% Hypertension 69% Pre-test Likelihood of CAD 58% FFR 0.80 32% Nitrates 99.6% Beta Blockers 78% Heart Rate (bpm) 63 Range 37-110 Prospective 54% mean dose (msv) 3 Retrospective 46% mean dose (msv) 14 Calcium score* Mean 302 >400 26% *Available for 214 patients
Sensitivity Discrimination of Ischemia* Per-Vessel (n=478) Per-Patient (n=251) FFR CT 0.93 CT 0.79 FFR CT 0.82 CT 0.63 1 - Specificity Greater discriminatory power for FFR CT versus CT stenosis Vessel (Δ 0.14, p<0.001) Patient (Δ 0.19, p<0.001) *Area under the receiver operating characteristics curve
Per-Patient Diagnostic Performance % 100 90 80 70 60 50 40 30 20 10 0 95% N=254 CI 92 86 81 79 65 Accuracy Specificity PPV Sensitivity NPV FFR CT < 0.80
Per-Patient Diagnostic Performance % 100 90 80 70 60 50 40 30 20 10 0 95% N=254 CI * * 81 79 64 53 51 46 40 34 65 94 91 92 93 Accuracy Specificity PPV Sensitivity NPV * 86 92 CT > 50% ICA > 50% FFR CT < 0.80 * p<0.0001
Per-Patient Diagnostic Performance % 100 90 20 10 0 95% N=254 CI * * 81 79 94 91 92 93 80 * 70 64 65 60 53 51 50 FFR 46 40 40 CT correctly reclassified 68% of 34 CT false positive to true negatives 30 Accuracy Specificity PPV Sensitivity NPV 86 92 CT > 50% ICA > 50% FFR CT < 0.80 * p<0.0001
Per-Vessel Diagnostic Performance % 100 90 80 70 60 50 40 30 20 10 0 N=484 95% C * * 86 86 71 65 66 * 60 40 33 61 83 84 84 92 94 Accuracy Specificity PPV Sensitivity NPV 95 CT > 50% ICA > 50% FFR CT < 0.80 * p<0.0001
Case Example LAD stenosis 70-90% FFR CT 0.93 FFR 0.94 FFR CT Model
Conclusions FFR CT met its primary endpoint in this trial and demonstrated high diagnostic accuracy for detection of ischemia When compared to anatomic interpretation by coronary CTA or invasive angiography, FFR CT led to a marked increase in diagnostic accuracy, specificity, and PPV FFR CT is performed from standard acquired CT datasets without the need for additional imaging, radiation or medication
Specificity Diagnostic performance of Coronary diagnostic tests for Functional (FFR 0.80) disease 100% FFR Gold Standard 90% 80% NXT FFR CT 70% 60% 50% 40% Stress Echo Jung, EHJ 2008 ccta Min, JAMA 2012 Koo, JACC 2011 Meijboom, JACC 2008 Norgaard 2013 SPECT Melikian, JACC CV Interventions 2010 Invasive Angiography Park, JACC CV Interventions 2012 Meijboom, JACC 2008 IVUS Waksman, JACC 2013 ccta TAG Yoon, JACC Imaging, 2012 FFR CT Norgaard 2013 30% 30% 40% 50% 60% 70% 80% 90% 100% Sensitivity
Thank you.