Author + information
- Received March 19, 2018
- Revision received August 16, 2018
- Accepted August 17, 2018
- Published online October 17, 2018.
- Gianluca Pontone, MD, PhDa,∗ (, )
- Andrea Baggiano, MDa,
- Daniele Andreini, MD, PhDa,b,
- Andrea I. Guaricci, MDc,d,
- Marco Guglielmo, MDa,
- Giuseppe Muscogiuri, MDa,
- Laura Fusini, MDa,
- Fabio Fazzari, MDe,
- Saima Mushtaq, MDa,
- Edoardo Conte, MDa,
- Giuseppe Calligaris, MDa,
- Stefano De Martini, MDa,
- Cristina Ferrari, MDa,
- Stefano Galli, MDa,
- Luca Grancini, MDa,
- Paolo Ravagnani, MDa,
- Giovanni Teruzzi, MDa,
- Daniela Trabattoni, MDa,
- Franco Fabbiocchi, MDa,
- Alessandro Lualdi, MDa,b,
- Piero Montorsi, MDa,b,
- Mark G. Rabbat, MDf,g,
- Antonio L. Bartorelli, MDa,h and
- Mauro Pepi, MDa
- aCentro Cardiologico Monzino, IRCCS, Milan, Italy
- bDepartment of Cardiovascular Sciences and Community Health, University of Milan, Milan, Italy
- cDepartment of Emergency and Organ Transplantation, Institute of Cardiovascular Disease, University Hospital Policlinico of Bari, Bari, Italy
- dDepartment of Medical and Surgical Sciences, University of Foggia, Foggia, Italy
- eDepartment of Cardiology, University Hospital P. Giaccone, Palermo, Italy
- fLoyola University of Chicago, Chicago, Illinois
- gEdward Hines Jr. VA Hospital, Hines, Illinois
- hDepartment of Biomedical and Clinical Sciences Luigi Sacco, University of Milan, Milan, Italy
- ↵∗Address for correspondence:
Dr. Gianluca Pontone, Centro Cardiologico Monzino, IRCCS, Via C. Parea 4, 20138 Milan, Italy.
Objectives This study sought to compare the diagnostic accuracy of coronary computed tomography angiography (cCTA) with that of cCTA+fractional flow reserve derived from cCTA datasets (FFRCT) and that of cCTA+static stress-computed tomography perfusion (stress-CTP) in detecting functionally significant coronary artery lesions using invasive coronary angiography (ICA) plus invasive FFR as the reference standard.
Background FFRCT and static stress-CTP are new techniques that combine anatomy and functional evaluation to improve assessment of coronary artery disease (CAD) using cCTA.
Methods A total of 147 consecutive symptomatic patients scheduled for clinically indicated ICA+invasive FFR were evaluated with cCTA, FFRCT, and stress-CTP.
Results Vessel-based and patient-based sensitivity, specificity, and negative predictive values, and positive predictive values, and accuracy rates of cCTA were 99%, 76%, 100%, 61%, 82%, and 95%, 54%, 94%, 63%, 73%, respectively. cCTA+FFRCT showed vessel-based and patient-based sensitivity, specificity, and negative predictive values, and positive predictive values and accuracy rates of 88%, 94%, 95%, 84%, 92%, and 90%, 85%, 92%, 83%, 87%, respectively. Finally, cCTA+stress-CTP showed vessel-based and patient-based sensitivity, specificity, and negative predictive values, and positive predictive values and accuracy rates of 92%, 95%, 97%, 87%, 94% and 98%, 87%, 99%, 86%, 92%, respectively. Both FFRCT and stress-CTP significantly improved specificity and positive predictive values compared to those of cCTA alone. The area under the curve to detect flow-limiting stenoses of cCTA, cCTA+FFRCT, and cCTA+CTP were 0.89, 0.93, 0.92, and 0.90, 0.94, and 0.93 in a vessel-based and patient-based model, respectively, with significant additional values for both cCTA+FFRCT and cCTA+CTP versus cCTA alone (p < 0.001) but no differences between cCTA+FFRCT versus cCTA+CTP.
Conclusions FFRCT and stress-CTP in addition to cCTA are valid and comparable tools to evaluate the functional relevance of CAD.
Coronary computed tomography angiography (cCTA) has been introduced as an alternative imaging modality to diagnose coronary artery disease (CAD) with low radiation exposure (1,2) and excellent prognostic assessment (3–6). However, there is concern regarding use of cCTA in the subset of patients who are at intermediate-to-high risk due to the limited positive predictive value of cCTA (7,8). In this regard, new techniques such as fractional flow reserve derived from cCTA datasets (FFRCT) (9–11) and stress computed tomography perfusion (stress-CTP) (12–15) recently emerged as potential strategies to combine anatomy and functional evaluations of CAD.
However, there are few data (16) for direct comparison among these techniques. Therefore, this study sought to compare the diagnostic accuracy of cCTA versus cCTA+FFRCT versus cCTA+stress-CTP to detect functionally significant coronary artery lesions by using invasive coronary angiography (ICA) plus invasive FFR as the reference standard.
Screening procedure and enrolment
The PERFECTION (PERfusion Versus Fractional Flow Reserve CT Derived In Suspected CoroNary) study was a longitudinal, prospective, consecutive cohort study designed to compare the feasibility and accuracy of integrated cCTA+FFRCT versus that of cCTA+stress-CTP for the diagnosis of functionally significant CAD (17). We screened consecutive symptomatic patients with suspected CAD referred for nonemergent, clinically indicated ICA between October 2015 and May 2017. From an original cohort of 928 patients, 781 were excluded according to the criteria shown in Figure 1, which included low to intermediate pre-test likelihood of CAD according to the updated Diamond-Forrester risk model score (18) (n = 82); prior clinically documented myocardial infarction (n = 40); history of surgical or percutaneous coronary artery revascularization (n = 415); suspicion of acute coronary syndrome (n = 21); need for an emergent procedure within 48 h of presentation (n = 13); evidence of clinical instability (n = 9); contraindications to contrast agents or impaired renal function (n = 47); inability to sustain a breathhold (n = 9); pregnancy (n = 0); atrial fibrillation or flutter (n = 40); body mass index (BMI) >35 kg/m2 (n = 26); presence of a pacemaker or implantable cardioverter-defibrillator (n = 34); contraindications to sublingual nitrates, beta-blockade, and adenosine (n = 42).
The institutional ethical committee study approved the protocol, and all patients meeting the selection criteria were asked to sign an informed consent. A structured interview was performed to collect a clinical history and cardiac risk factors.
Figure 2 shows the study protocol. Patients were asked to refrain from smoking and caffeine for 24 h and to maintain fasting for 6 h before the scan. In patients with a resting heart rate (HR) >65 beats/min, metoprolol was intravenously administered, with a titration dose up to 15 mg to achieve a target HR of ≤65 beats/min. Before the rest scan, all patients received sublingual nitrates to ensure coronary vasodilation.
REST-cCTA performance and interpretation
Rest-cCTA was performed using a Revolution CT scanner (GE Healthcare, Milwaukee, Wisconsin) according to the recommendations of the Society of Cardiovascular Computed Tomography (SCCT) (19), using the following parameters: slice configuration of 256 × 0.625 mm with scintillator detector (Gemstone detector, GE Healthcare); gantry rotation time of 280 ms; tube voltage of 120 KVp and 100 KVp in patients with BMI >30 kg/m2 and ≤30 kg/m2, respectively; and effective tube current of 500 mA. One-beat axial scan was used in all patients, with variable padding ranging between 70% to 80% and 40% to 80% of cardiac cycle in patients with HR <65 beats/min and those with ≥65 beats/min, respectively. All patients received a 70-ml bolus of iodixanol 320 (320 mg/ml, Visipaque, GE Healthcare) at an infusion rate of 6.2 ml/s, followed by 50 ml of saline solution. All scans were performed using the bolus tracking technique by using visual assessments to determine timing of image acquisition. A new generation post-processing adaptive statistical iterative reconstruction algorithm (ASIR-V, GE Healthcare) was used instead of the standard filtered back-projection algorithm. Datasets of cCTA were transferred to an image-processing workstation (Advantage Workstation version 4.7, GE Healthcare) to perform quantitative coronary analysis according to SCCT guidelines for reporting (19). Two cardiac imagers who were blinded to clinical history, stress-CTP, and invasive evaluation findings independently evaluated the reconstructed images. For analysis of cCTA, coronary arteries were segmented as suggested by the American Heart Association (AHA) (20). Causes of artifacts and image quality evaluation with a Likert score were established as previously described (8). In each coronary artery, coronary atherosclerosis was defined as the presence of any tissue structures larger than 1 mm2, either within the coronary artery lumen or adjacent to it that could be discriminated from the surrounding pericardial tissue, epicardial fat, or vessel lumen itself. The severity of coronary lesions was quantified in multiplanar curved reformatted images by identifying the minimum diameter and reference diameter of all stenoses and quantified according to SCCT guidelines (19). Stenosis >50% was considered significant from an anatomical point of view. A third cardiac imager adjudicated the scores in cases of disagreement.
FFRCT performance and interpretation
All cCTA datasets were sent to HeartFlow (Redwood City, California) for FFRCT analysis as previously described (11,21). An FFRCT <0.80 was considered significant.
Static STRESS-CTP performance and interpretation
Vasodilation was induced by IV administration of adenosine (0.14 mg/kg/min over 4 min). At the end of the third minute of adenosine infusion, a single data sample was acquired during first-pass enhancement of cCTA by using the same protocol described for rest-cCTA (Figure 2). All datasets of static stress-CTP were transferred to an image-processing workstation (Advantage Workstation Version 4.7, GE Healthcare), and 2 cardiac imagers blinded to clinical history, rest cCTA, and invasive evaluation findings independently evaluated the reconstructed images. The myocardial wall was evaluated on short-axis (apical, mid, and basal slices) and long-axis views (2-, 3- and 4-chamber projections) with 4-mm-thick average multiplanar reformatted images in diastolic phases (70% and 80% of cardiac cycle). A narrow window width and level (350 W and 150 L) was used for perfusion defect evaluation. Each myocardial segment was correlated to the specific coronary territory according to a modified AHA classification as described by Cerci et al.(22). In detail, the entry criterion for the algorithm was the presence of both at least 1 coronary arterial lesion >50% diameter stenosis and at least 1 myocardial perfusion defect. For each vessel, the following territories were identified: 1) primary territory consisting of myocardial territories in which blood flow is supplied by the coronary vessel in the most common right dominant anatomic coronary pattern; 2) secondary territories consisting of myocardial territories for which blood flow may be supplied by the coronary vessel under some normal anatomic variations that need confirmation; and 3) tertiary territories consisting of myocardial territories where blood flow is usually not supplied by the coronary vessel. The adjudication process was applied each time there was a coronary arterial lesion with >50% diameter stenosis and at least 1 myocardial perfusion defect in the secondary territories. After myocardial segmentation, a 4-point image quality score was assigned to each myocardial segment regarding the diagnostic confidence of perfusion defect evaluation 1 = very uncertain (i.e., poor confidence; could be an artifact or poor image quality); 2 = uncertain (i.e., moderate confidence, probably an artifact and less likely a perfusion defect); 3 = rather certain (i.e., good confidence, probably a defect, good image quality/no or minor artifacts); and 4 = very certain (i.e., excellent image quality/no artifacts). Perfusion defects were defined as subendocardial hypoenhancements encompassing ≥25% of transmural myocardial thickness within a specific coronary territory.
ICA and invasive FFR performance and interpretation
In all patients, certified interventional cardiologists performed ICA within 60 days after the cCTA examination according to the American College of Cardiology/AHA Task Force on Practice Guidelines and the Society for Cardiac Angiography and Interventions (23). Coronary angiograms were analyzed at the clinical site by an interventional cardiologist blinded to cCTA, stress-CTP, and FFRCT findings. Severity of luminal narrowing was assessed using the same semiquantitative score previously described for cCTA. Coronary artery stenoses ≥80% or totally occluded vessels were considered functionally significant without performing invasive FFR measurements, although all stenoses ranging between 30% and 80% were evaluated by clinically indicated invasive FFR (24,25). For FFR, the pressure wire (Certus pressure wire; St. Jude Medical Systems, St. Paul, Minnesota) was calibrated and electronically equalized using the aortic pressure before being placed distal to the stenosis in the distal third of the coronary artery being interrogated. Glyceryl trinitrate (100 mg) was injected intracoronary to prevent vasospasm. Adenosine was administered (140 μg/kg/min) intravenously. At steady-state hyperemia, FFR was assessed using the RadiAnalyzer Xpress (Radi Medical Systems, Uppsala, Sweden), calculated by dividing the mean coronary pressure, measured with the pressure sensor placed distal to the stenosis, by the mean aortic pressure measured through the guide catheter. Intermediate stenoses ≤0.8 found by invasive FFR or stenoses with >80% diameter reduction or total occlusions were considered functionally significant.
For cCTA, the dose-length product, defined as total radiation energy absorbed by the patient body, was measured in mSv/mGy·cm. The effective radiation dose was calculated as the product of dose-length product times a conversion coefficient for the chest (K = 0.014 mSv/mGy·cm). For ICA, the effective radiation dose was calculated by multiplying the dose-area product by a conversion factor (K = 0.21 mSv/mGy·cm2) for lateral and posteroanterior radiation exposure in the chest area.
Statistical analysis was performed using a dedicated SPSS version 21.0 software (SPSS, Chicago, Illinois) and R version 2.15.2 ((R Foundation for Statistical Computing, Vienna, Austria). The sample size was estimated assuming a 30% prevalence of functionally significant CAD and diagnostic accuracy for FFRCT in a vessel-based model of 86%. A sample size of 150 patients, corresponding to 450 vessels, was considered powered to detect a difference of 4% between FFRCT and stress myocardial CTP at a significance level of 5% and at least 90% power, using a 2-sided test. Continuous variables were expressed as mean ± SD and discrete variables as absolute numbers and percentages. The Spearman correlation and Bland-Altman analyses were used for comparing FFRCT to invasive FFR values. The chi-squared test or Fisher exact test was used to study differences in categorical data. For cCTA and integrated cCTA+FFRCT or cCTA+CTP protocols, the overall evaluability (i.e., number of evaluable coronary artery segments-to-all coronary artery segments ratio), sensitivity, specificity, negative and positive predictive values, and accuracy were calculated compared to those of ICA+ invasive FFR as reference standards. Specifically, the integrated evaluation was performed according to the following interpretation: A) non-obstructive CAD with negative matched functional evaluation was considered negative; B) nonobstructive CAD with positive matched functional evaluation was considered still negative; C) obstructive CAD with negative matched functional evaluation was considered negative; and D) obstructive CAD with positive matched functional evaluation was deemed positive. The McNemar test was used to calculate differences in terms of sensitivity, specificity, negative predictive value, positive predictive value, and accuracy; and area under the receiver operating characteristics (AUC) curves for each model was measured and compared using the DeLong method.
Table 1 lists patients’ clinical characteristics. All patients underwent ICA, and invasive FFR was measured in 98 of 147 patients (67%). Obstructive CAD was observed in 94 of 147 patients (64%), whereas the prevalence of functionally significant CAD was detected in 66 of 147 patients (45%). Among 122 vessels with functionally significant CAD, 53 (43%) were left anterior descending arteries, 32 (26%) were left circumflex coronary arteries, and 37 (30%) were right coronary arteries.
Image quality and overall evaluability of REST-cCTA
Rest-cCTA was successfully performed in all patients. The mean Likert score was 3.6 ± 0.8, and the overall evaluability of native coronary arteries was 98% (Online Table 1).
Image quality and overall evaluability of FFRCT
FFRCT was successfully performed in 143 of 147 patients (98%). The analysis was rejected in the remaining 4 patients for motion artifacts.
Image quality and overall evaluability of STRESS-CTP
Stress-CTP was successfully performed in 144 of 147 patients with a mean HR during the scan of 77.5 ± 14.1 beats/min (Table 1). The stress phase was interrupted because of dyspnea in 2 patients and chest pain in 1 patient. In most myocardial segments, myocardial perfusion interpretation was classified as very certain or rather certain (61% and 22%, respectively), whereas it was classified as very uncertain in only 1%. The mean image quality score for myocardial perfusion was 3.4 ± 0.3 (Online Table 2).
Diagnostic accuracy of REST-cCTA
The diagnostic performance of rest-cCTA, compared to ICA+invasive FFR, is listed in Table 2. Rest-cCTA demonstrated a vessel-based and patient-based sensitivity, specificity, negative predictive value, positive predictive value and diagnostic accuracy of 99%, 76%, 100%, 61%, 82% and 95%, 54%, 94%, 63%, 73%, respectively (Table 2).
Diagnostic accuracy of cCTA+FFRCT
There was good direct correlation of per-vessel FFRCT to invasive FFR (Pearson’s correlation coefficient 0.69; p< 0.001), with a slight underestimation of FFRCT compared with FFR (mean difference: 0.02 ± 0.13). The vessel-based and patient-based sensitivity, specificity, negative predictive value, positive predictive value and diagnostic accuracy of integrated cCTA+FFRCT were 88%, 94%, 95%, 84%, 92% and 90%, 85%, 92%, 83%, 87%, respectively (Table 2).
Diagnostic accuracy of REST-cCTA+STRESS-CTP
The diagnostic performance of the integrated cCTA+stress-CTP protocol is listed in Table 2. The vessel-based and patient-based sensitivity, specificity, negative predictive value, positive predictive value and diagnostic accuracy were 92%, 95%, 97%, 87%, 94% and 98%, 87%, 99%, 86%, 92%, respectively (Table 2).
Comparison between integrated protocols of cCTA alone, cCTA+FFRCT, and cCTA+STRESS CTP
The AUCs to detect flow-limiting stenosis of cCTA, cCTA+FFRCT and cCTA+CTP were 0.89, 0.93, 0.92 and 0.90, 0.94, 0.93 in vessel- and patient-based models, respectively (Figure 3), with significant additional values of both cCTA+FFRCT and cCTA+CTP versus cCTA alone (p<0.001) but no differences between cCTA+FFRCT versus cCTA+CTP. Figures 4 and 5⇓⇓ show representative cases of comparison between cCTA+stress-CTP and cCTA+FFRCT.
The main findings of this study are that both FFRCT and stress-CTP provide additional value in terms of specificity, positive predictive value, and diagnostic accuracy compared to rest-cCTA, without significant difference between these 2 approaches.
Several studies demonstrated that optimal medical therapy alone has efficacy similar to revascularization when obstructive CAD is not associated with ischemia. Therefore, accurate identification of patients with ischemic CAD is of major clinical importance. FFRCT demonstrated a 68% reduction of false positive cCTA cases in the NxT (Analysis of Coronary Blood Flow Using CT Angiography: Next Steps) trial (11), and regarding stress-CTP, several prospective studies assessed the diagnostic performance of a rest+stress-CTP protocol versus ICA plus invasive FFR using both static (12–15,26,27) and dynamic techniques (28–30).
Despite the fact that FFRCT and stress-CTP share good overall diagnostic performance, both have strengths and weaknesses. FFRCT does not require additional scan time and use of stressors and, therefore, is associated with low radiation exposure. However, it is based on several geometric pathophysiological assumptions. On the other hand, stress-CTP is potentially more representative of the ischemic cascade but requires additional scan time and use of a stressor agent and is associated with higher radiation exposure. In addition, FFRCT cannot be used to evaluate vessel stenosis in case a stent is present, whereas stress-CTP has no restriction in this regard. There are only 2 previous studies comparing the 2 techniques (16,31). Yang et al. (16) compared FFRCT to stress-CTP in 72 consecutive patients and found no significant differences between the AUC values of the 2 techniques (p = 0.84) that, however, were higher than those of cCTA alone (AUC: 0.919; p = 0.004; and AUC: 0.913; p = 0.004; respectively). However, this study is limited by retrospective design and the use of a dual-source 128-slice scanner that, despite its excellent temporal resolution, does not allow for single-beat acquisition. Similarly, Coenen et al. (31) showed that both FFRCT and stress-CTP have increased AUC compared to cCTA alone (0.78 for both techniques) and found that stress-CTP performed better than FFRCT (AUC 0.85). However, their sample size was small (72 patients) and underpowered to evaluate differences between the 2 techniques, and they used a stress-CTP protocol based on dynamic acquisition with a 128-slice scanner that is associated with high radiation exposure.
Our study confirms that both FFRCT and stress-CTP provide additional diagnostic value compared to rest-cCTA and that there are no significant differences in term of diagnostic performance alone. However, our results have some strengths. First, the study design is prospective. Second, the target study population was at intermediate-to-high risk for CAD. This is the ideal setting for the use of additional functional testing with CT. Indeed, a high pre-test likelihood of CAD is associated with an increased burden of calcified atherosclerotic disease that impairs the value of cCTA to correctly rule out CAD (8,32). Third, the whole-heart CT scanner technology used in this study enables isophasic, single-beat imaging of the entire coronary tree, and it could be particularly suitable for static CTP (33). Finally, our choice of ICA plus invasive FFR as the reference standard was considered state of art for validation studies.
Despite those strengths, in our study, we did not find differences between FFRCT and CTP and their clinical applications could sometimes be different based on their strengths and weaknesses. For example, in case of standard 64-slice scanner technology, FFRCT could be preferred to stress-CTP. Indeed, the acquisition of data from sequential heartbeats affects the attenuation gradient and may result in a heterogeneous iodine distribution, mimicking perfusion defects. Similarly, in case of perfusion defect that does not match with obstructive CAD, a beam-hardening artifact should be taken into account, and the addition of FFRCT could be useful. Finally, in case of 3-vessel disease, CTP may not unmask balanced ischemia that could be better detected by FFRCT. On the other hand, CT-FFRct uses the cCTA images as boundary conditions for the computational fluid dynamic analysis of the coronary tree, and therefore, the technique is sensitive to factors that result in artifacts of the underlying coronary artery images, such as motion artifact or significant coronary calcification that usually does not affect the performance of CTP.
Our study has limitations. First, we used static stress-CTP rather than dynamic stress-CTP. Therefore, a quantitative analysis of myocardial perfusion was not feasible (34). Second, invasive FFR was not performed in all vessels but only in those with intermediate lesions. Considering that a significant percentage of coronary artery stenosis >80% is associated with a normal, invasive FFR, a potential overestimation of functionally significant CAD could be occurred. However, considering that the same reference standard was used for both FFRCT and stress-CTP, this limitation should have minimal impact on the comparison of diagnostic accuracy of these two techniques. Third, we used narrow exclusion criteria, and our results are therefore limited to this populations, with the same prevalence of functionally significant disease. Further studies should test these techniques in the general population. Fifth, according to our standard clinical practice, we did not perform calcium score before cCTA. Therefore, no subanalysis can be performed in terms of impact of calcium score on FFRCT compared to stress CTP performance. Finally, it is noteworthy that perfusion techniques are sensitive to both epicardial vessel obstruction and microvascular disease, whereas FFRCT and invasive FFR are only able to assess epicardial lesions and vessel-specific ischemia.
The addition of both FFRCT and stress-CTP to cCTA is a valid and feasible strategy to evaluate the functional relevance of CAD. Based on these results, in most patients with suspected CAD, cCTA alone and integrated with FFRCT or CTP is a robust tool to diagnose functionally relevant stenoses.
COMPETENCY IN MEDICAL KNOWLEDGE: Coronary computed tomography angiography (cCTA) has been introduced as an alternative imaging modality to diagnose CAD with low radiation exposure and excellent prognostic assessment. However, there is concern regarding cCTA use in the subset of patients at intermediate-to-high risk due to the limited positive predictive value of cCTA, particularly in the presence of calcified coronary lesions. In this regard, new techniques such as fractional flow reserve derived from cCTA datasets (FFRCT) and stress computed tomography perfusion (stress-CTP) recently emerged as potential strategies to combine anatomy and functional evaluation of CAD in a “one-stop-shop.” The aim of this study was to compare the diagnostic accuracy of cCTA versus cCTA+FFRCT versus cCTA+stress-CTP in detecting functionally significant coronary artery lesions in consecutive symptomatic patients at intermediate-to-high risk for CAD using ICA with invasive FFR as reference standard. We found that both FFRCT and stress-CTP provides additional value in terms of specificity, positive predictive value and diagnostic accuracy when compared to rest-cCTA with no statistically significant difference between them.
TRANSLATIONAL OUTLOOK: Based on these results, in most patients with suspected CAD, cCTA alone is sufficient to exclude functionally relevant CAD when an obstructive CAD is absent. On the contrary, in the setting of obstructive CAD, both FFRCT or CTP are equally accurate to detect functionally significant stenosis.
The authors thank Dr. Alessandra Terragni, Clinical Trial Office, Centro Cardiologico Monzino, IRCCS, Milan, Italy.
Supported by the Current Research Program of Italian Ministry of Health, Rome, Italy. Dr. Pontone has received research support and speaker honoraria from General Electric Health, Bracco, Medtronic, Bayer, and Heartflow. Dr. Andreini has received research support through his institution; and speaker honoraria from General Electric Health, Bracco, and Heartflow. All other authors have reported that they have no relationships with industry relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- American Heart Association
- adaptive statistical iterative reconstruction algorithm
- area under the curve
- coronary artery disease
- coronary computed tomography angiography
- fractional flow reserve derived from cCTA datasets
- heart rate
- invasive coronary angiography
- Society of Cardiovascular Computed Tomography
- stress-computed tomography perfusion
- Received March 19, 2018.
- Revision received August 16, 2018.
- Accepted August 17, 2018.
- 2018 American College of Cardiology Foundation
- Pontone G.,
- Andreini D.,
- Bartorelli A.L.,
- et al.
- Min J.K.,
- Dunning A.,
- Lin F.Y.,
- et al.
- Pontone G.,
- Andreini D.,
- Bartorelli A.L.,
- et al.
- Hoffmann U.,
- Ferencik M.,
- Udelson J.E.,
- et al.
- Williams M.C.,
- Hunter A.,
- Shah A.S.V.,
- et al.
- Dewey M.,
- Vavere A.L.,
- Arbab-Zadeh A.,
- et al.
- Koo B.K.,
- Erglis A.,
- Doh J.H.,
- et al.
- Norgaard B.L.,
- Leipsic J.,
- Gaur S.,
- et al.
- Sara L.,
- Rochitte C.E.,
- Lemos P.A.,
- et al.
- Pontone G.,
- Andreini D.,
- Guaricci A.I.,
- et al.
- Yang D.H.,
- Kim Y.H.,
- Roh J.H.,
- et al.
- Pontone G.,
- Andreini D.,
- Guaricci A.I.,
- et al.
- Taylor C.A.,
- Fonte T.A.,
- Min J.K.
- Cerci R.J.,
- Arbab-Zadeh A.,
- George R.T.,
- et al.
- Levine G.N.,
- Bates E.R.,
- Blankenship J.C.,
- et al.
- Bettencourt N.,
- Chiribiri A.,
- Schuster A.,
- et al.
- Wong D.T.,
- Ko B.S.,
- Cameron J.D.,
- et al.
- Coenen A.,
- Rossi A.,
- Lubbers M.M.,
- et al.
- Ladeiras-Lopes R.,
- Bettencourt N.,
- Ferreira N.,
- et al.
- Techasith T.,
- Cury R.C.