PR O G RE S S I N C ARDI O V A S CU L A R D I S EA S E S XX ( 2 0 1 5) XXX – XXX

Available online at www.sciencedirect.com

ScienceDirect www.onlinepcd.com

CT Assessment of Myocardial Perfusion and Fractional Flow Reserve Edward Hultena, b , Amir Ahmadic , Ron Blanksteina,⁎ a

Non-Invasive Cardiovascular Imaging Program, Departments of Medicine and Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA b Cardiology Service, Department of Medicine, Walter Reed National Military Medical Center and Uniformed Services University of Health Sciences, Bethesda, MD c Division of Cardiology, University of British Columbia, Vancouver, Canada

A R T I C LE I N F O

AB ST R A C T

Keywords:

Coronary computed tomography angiography (CTA) offers a non-invasive method to detect

Computed tomography

coronary plaque and stenosis. However, to date, CTA has been most useful as a method of

Coronary

ruling out coronary artery disease (CAD) among patients with low to intermediate pretest

Coronary artery disease

probability of significant CAD. The reduced specificity of CTA for detecting physiologically

Atherosclerosis

significant stenosis is a known limitation of this technique, particularly since some patients

Myocardial perfusion imaging

require additional functional testing following CTA. Therefore, intense interest has focused on the development of methods to determine the functional significance of anatomical lesions identified by CTA. This article will discuss two emerging methods: stress myocardial perfusion imaging using CT, or CT perfusion, and computer simulation of fractional flow reserve. © 2015 Elsevier Inc. All rights reserved.

Coronary computed tomography (CT) angiography (CTA) emerged in the 1990s as an experimental non-invasive method of diagnosing coronary artery disease (CAD).1 Due to rapid evolution and technological advances in multi-detector CT as well as image reconstruction methods, CTA quickly gained clinical acceptance as a diagnostically accurate2,3 anatomic test with important prognostic value.4 However, to date, CTA has been most useful as a method of ruling out CAD among patients with low to intermediate pretest probability of significant CAD.5 The reduced specificity of CTA for detecting physiologically significant stenosis continues to be a concern, particularly since some patients require additional functional testing following CTA.

The rationale for evaluating ischemia Even though the degree of stenosis is one of the factors that partially predicts the presence or absence of lesion specific ischemia, the association between coronary anatomy and ischemia is poor. For instance, patients can have no ischemia in presence of significant stenosis (NIPSS) and presence of ischemia with no severe stenosis (PINSS).6,7 An ideal coronary diagnostic test would possess the ability to accurately assess the burden of atherosclerosis as well as to determine the physiological consequence of anatomical lesions by determining the presence and severity of ischemia. Therefore, intense interest has focused on the development of methods

Statement of Conflict of Interest: see page 8. ⁎ Address reprint requests to Ron Blankstein, MD, Non-Invasive Cardiovascular Imaging Program, Departments of Medicine (Cardiovascular Division) and Radiology, Brigham and Women's Hospital, 75 Francis St, Boston, MA 02115. E-mail address: [email protected] (R. Blankstein). http://dx.doi.org/10.1016/j.pcad.2015.03.003 0033-0620/© 2015 Elsevier Inc. All rights reserved.

Please cite this article as: Hulten E, et al. CT Assessment of Myocardial Perfusion and Fractional Flow Reserve. Prog Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.pcad.2015.03.003

2

P R O GRE S S I N CA RDI O VAS CU L AR D I SE AS E S XX ( 2 0 15 ) XXX– X XX

Abbreviations and Acronyms

to determine the functional significance of ALARA = as low as reasonably anatomical lesions idenachievable (radiation exposure) tified by CTA. CAD = coronary artery disease Coronary angiography has been used as CT = computed tomography clinical gold standard CTA = computed tomography for detecting CAD for angiography the past few decades and the degree of anaCTP = computed tomography tomical stenosis has perfusion shown to have a powerFFR = fractional flow reserve ful prognostic value.8,9 However, treatment straFFRCT = computer simulation of tegies based on angiofractional flow reserve based graphic findings and upon computed tomography severity of anatomical ICA = invasive coronary stenosis alone do not angiography result in reduction of major adverse cardiac MACE = major adverse cardiac outcomes outcomes (MACE) compared to optimal mediMDCT = multi-detector cal therapy (OMT).10 computed tomography Unlike anatomy driven NIPSS = no ischemia in presence treatment strategies, of significant stenosis physiology based treatment strategies have MPI = myocardial perfusion been shown to be assoimaging ciated with improved MRI = magnetic resonance While outcomes.11–16 imaging most non-invasive techniques for detecting isPINSS = presence of ischemia chemia provide a global with no severe stenosis assessment of myocarSPECT = single photon emission dial blood flow (MBF), computed tomography invasive assessment with fractional flow reserve (FFR) offers a lesion specific measure of ischemia determined by calculating the ratio of flow across a lesion to a flow at maximum hyperemia in a presumable absence of stenosis. The fractional flow reserve versus angiography for guiding percutaneous coronary intervention (FAME) and FAME 2 studies demonstrated that FFR guided therapy is superior to both angiography guided therapy and OMT, respectively. 13–16 This article will discuss two emerging methods of using cardiac CT to assess ischemia: stress myocardial perfusion imaging using CT, or CT perfusion (CTP), and computer simulation of fractional flow reserve (FFRCT).

CT myocardial perfusion imaging CT scans of canine hearts undergoing stress CTP demonstrated in 2007 that CTP derived quantitative perfusion correlated well with myocardial perfusion determined by microspheres.17 The first reports of CTP as technically feasible and accurate in human clinical studies were reported in two separate studies by Blankstein and George in 2009.18,19 Since that time, the

diagnostic accuracy of CTP has been studied in comparison to a variety of reference standards, including myocardial perfusion imaging (MPI) by single photon emission CT (SPECT) or magnetic resonance imaging (MRI), invasive or noninvasive angiographic stenosis, and invasive FFR.20–26

CTP protocols In order to assess both coronary anatomy and myocardial ischemia using CTP, two separate CT acquisitions are required: one for vasodilator induced stress MPI and one for rest MPI and coronary CTA. However, a current challenge for the field of CTP is a lack of standardized approaches for image acquisition and post-processing. While close attention to heart rate is required for the rest acquisition, heart rate control is not as stringent for stress perfusion MPI since the coronaries will not be interpreted on this study. A delay ranging from 10 to 30 minutes is currently employed between acquisitions to allow for CT contrast washout and/or reversal of the effects of the pharmacological vasodilator used. For the stress perfusion, peak hyperemia may be achieved with adenosine, dypyridamole, or regadenoson. Regadenoson has the benefit of not requiring a continuous intravenous infusion pump since it is administered as a single injection that results in several minutes of hyperemia. A clinical example of CTP is demonstrated in Fig 1. A common question is whether to employ a stress–rest perfusion protocol or a rest–stress protocol. Investigators have used either protocol with similar clinical result, but for practical workflow a rest–stress protocol would have the advantage that only patients who are identified to have a possible stenosis would then undergo stress perfusion imaging. Routine use of stress and rest CTP along with CTA for all patients is unlikely to be cost-effective since only approximately one third4 of patients currently referred to CTA have possible stenosis. Fig 2 illustrates the general protocol for stress–rest (top panel) and rest–stress for CTP. Nearly any modern scanner with 64 or more detectors can accurately evaluate CT MPI, although there are unique considerations for each specific CT platform. First of all is the z-axis coverage, which is mostly determined by the number of detectors. While the first reported CT MPI studies used 64 multi-detector CT (MDCT) platforms, which require image acquisition over multiple heart beats, the use of this technique results in myocardial contrast enhancement that is not homogeneous due to the slight differences in acquisition time of each slab. Such temporal differences are usually not significant enough to impair a qualitative assessment of CTP, but may be problematic when quantitative techniques are used or when assessing for differences in contrast opacification gradients. With 320 MDCT, sufficient z-axis coverage allows for a single beat acquisition that provides more homogeneous MPI data. Another important consideration is the estimated effective radiation dose delivered to the patient. Most patients who would clinically benefit from the additive information of CT MPI are older and have CAD and therefore the affect of medical radiation exposure upon lifetime risk of malignancy will be negligible. However, public health guidelines including recommendations of the American Heart Association27 recommend to reduce

Please cite this article as: Hulten E, et al. CT Assessment of Myocardial Perfusion and Fractional Flow Reserve. Prog Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.pcad.2015.03.003

3

PR O G RE S S I N C ARDI O V A S CU L A R D I S EA S E S XX ( 2 0 1 5) XXX – XXX

Test Bolus

Stress Perfusion Protocol CT Stress Perfusion Protocol Scout Images

Stress

Stress Perfusion

Rest Perfusion

Vasodilator Contrast

Contrast

Rest Perfusion

Stress Perfusion

Test Bolus

Rest

Scout Images

Start

Contrast

Vasodilator Contrast

Start Fig 2 – Example of stress–rest (top panel) and rest–stress (lower panel) CT perfusion protocols.

Fig 1 – Example of stress induced CT perfusion defect involving the anterior, anteroseptal, and lateral walls (top image). The rest perfusion images demonstrated a small defect in the anterior wall. Collectively, these findings were consistent with severe ischemia and the patient was found to have severe stenosis of the left anterior descending and left circumflex coronary arteries.

myocardial perfusion, multi planar reformats (MPR) images are viewed using thick slices (6–8 mm) and using average intensity projection of each slab (as opposed to maximum intensity projection which is used for coronary imaging). Finally, the grayscale window width and level are set at approximately 300/150.30 Table 1 summarizes the strengths and limitations of CTP.

Highlights of studies investigating CTP imaging

medical radiation by the “as low as reasonably achievable” (ALARA) principle, particularly for cardiac patients who may receive cumulative radiation doses due to serial imaging.28 Differences in scanner technology and acquisition protocol have a significant impact on the estimated effective radiation dose.29 A lower dose can be achieved by using axial acquisition with prospective ECG gating, as well as reducing the tube voltage (kV) in non-obese patients. While various semi-quantitative CTP techniques exist, most of these rely on acquiring multiple data sets over time. Such dynamic CTP protocols are associated with a higher dose than when only a single data set is acquired. However, when only a single data set is acquired during first-pass stress or rest perfusion, it is essential to ensure that the image acquisition is appropriately timed and occurs during the initial wash-in of contrast into the myocardium. This can be achieved with either a test bolus or bolus tracking technique. Fig 3 depicts the estimated effective radiation dose delivered in several CTP studies. Following raw data acquisition, CTP images should be reconstructed with a smooth filter, and using beam hardening correction, when available. For optimal image display of

Initial animal studies demonstrated that quantitative myocardial perfusion by CTP correlated with microspheres in six dogs.17 Subsequently, single center studies on 64 MDCT were reported first by Blankstein et al. and George et al. in 2009.18,19 Blankstein et al. demonstrated a per-vessel sensitivity and specificity of 93% and 74% for CTA-CTP in comparison to invasive coronary angiography (ICA)-SPECT.6 George et al. evaluated CTP in 40 subjects and determined that the sensitivity and specificity of CTA-CTP in comparison to ICA-SPECT were 86% and 92%, in the per-patient analysis and 79% and 91% in a per-vessel analysis.7 Following these encouraging initial studies, several additional single center, observational studies evaluated CTP using a variety of different techniques with encouraging accuracy results in comparison to both invasive angiography and functional imaging.20–26 Not all studies reported accuracy according to CTA and CTP in comparison to a reference test, but Ko et al. reported among 40 symptomatic patients that the accuracy of >50% coronary stenosis as determined by CTA for detection of FFR significance (FFR 50% stenosis by CTA with ischemia determined by CTP to 87% sensitivity, 95% specificity, and 93% accuracy.26

Please cite this article as: Hulten E, et al. CT Assessment of Myocardial Perfusion and Fractional Flow Reserve. Prog Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.pcad.2015.03.003

4

P R O GRE S S I N CA RDI O VAS CU L AR D I SE AS E S XX ( 2 0 15 ) XXX– X XX

Table 1 – Strengths and limitations of CT perfusion. Strengths: Simultaneous data on coronary anatomy and perfusion Higher accuracy than other SPECT MPI for detecting of multi-vessel disease Fast and well tolerated exam Limitations: Artifacts (motion; beam hardening) Lower contrast resolution than SPECT or cardiac MRI MPI Ionizing radiation Large contrast dose Requires optimal contrast timing No outcome data

Fig 3 – Estimated effective radiation dose associated with various CT perfusion protocols. Reprinted from Hulten et al. JNC. 2012;19:588–600.

Following these single center studies, more recently the multicenter, international study Coronary Artery Evaluation using 320-row Multi-detector Computed Tomography Angiography and Myocardial Perfusion (CORE 320) was conducted. This study evaluated 381 patients who underwent CTP and CTA using the 320 MDCT, as well as SPECT MPI and invasive angiography. When compared to the reference standard of a 50% stenosis by ICA, which was also associated with a perfusion defect by SPECT, the presence of stenosis by CTA ≥ 50% demonstrated a sensitivity of 94% and specificity of 64%. However, when CTP was added to CTA, there was a modest improvement in specificity, with larger and more severe perfusion defects having the highest specificity.31 When evaluating the area under the receiver operating characteristic (ROC) curve, the presence of stenosis by CTA had an area under the curve (AUC) of 0.82 (0.78–0.85) which improved to 0.87 (0.84–0.89) with the addition of CTP. There was an even greater improvement among patients without known prior CAD to an AUC of 0.93 (0.89–0.97). In a separate analysis, directly comparing the accuracy of CTP and SPECT MPI to identify 50% stenosis by invasive angiography, CTP demonstrated improved performance compared to SPECT MPI, a finding that was driven in part by the higher sensitivity for left main and multi-vessel CAD.32 In addition to the CORE 320, a subsequent multicenter trial has been conducted using regadenoson for CTP in comparison with SPECT. This trial incorporated a variety of different vendors' CTA hardware and included sites that did not have any prior experience in CTP.33,34 Among 110 patients, this study demonstrated that CTP had good (87%) agreement with SPECT and that the accuracy of CTA (0.69) for detecting a reversible perfusion defect on SPECT could be improved to 0.85 with the use of CTP. Regadenoson stress CTP met the predefined criteria of this study for noninferiority to SPECT. Finally dual-energy CT may offer an opportunity to further visualize myocardial perfusion or to help mitigate artifacts such as beam hardening that currently limit the accuracy of CTP. While initial data on the use of this technique have not demonstrated a definite, consistent benefit,35 an ongoing

multi-center trial investigating dual energy CTP has recently been initiated with ongoing enrollment: the dual-energy CT for ischemia determination compared to “gold standard” non-invasive and invasive techniques (DECIDE-Gold).36 Although these initial single center accuracy studies and multi-center trials have demonstrated the potential for CT to serve as an accurate “one-stop shop” for both coronary anatomy and MPI, the technique requires appropriate patient selection to ensure that it is performed in patients with a sufficient burden of CAD to warrant the additional time, cost, and radiation invested. Thus, CTA-CTP, while encouraging for patients with hemodynamically indeterminate coronary lesions, should not be a default imaging modality for patients without known CAD since many patients may be evaluated more simply by CTA alone. The interpretation of CTP images does require experience and knowledge in how to distinguish true perfusion defects from various artifacts (e.g. beam hardening or motion artifacts). Despite multiple studies showing the accuracy of CTP, including 2 multicenter studies, further data are needed regarding the clinical utility of CTP to benefit patient management and determine prognosis.

CT assessment of FFR While there is an ongoing debate among cardiologists as to whether coronary anatomy or physiology is more important in determining prognosis and impacting treatment decisions, these two techniques are often complimentary. Recognizing this, several advances in non-invasive imaging have allowed us to evaluate both coronary anatomy and physiology. One such advance is the FFRCT, which uses the dataset obtained from CTA to provide a lesion specific estimation of FFR, but without any additional radiation exposure or administration of any vasodilators.

FFRCT: the fundamentals Recent advances in the field of computational fluid dynamics (CFD) and advanced imaging based modeling have allowed for non-invasive calculation of FFR based on static CT images. As applied to typically acquired coronary CT angiograms, these technologies enable non-invasive calculation of FFR without

Please cite this article as: Hulten E, et al. CT Assessment of Myocardial Perfusion and Fractional Flow Reserve. Prog Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.pcad.2015.03.003

PR O G RE S S I N C ARDI O V A S CU L A R D I S EA S E S XX ( 2 0 1 5) XXX – XXX

modification of image acquisition protocols, additional imaging, radiation, or added medications for vasodilation. The details of methods used in calculation of FFRCT have been described elsewhere.37,38 In general, by using the “Navier– Stokes Equations” of CFD and based on the relationship of mass conservation and momentum balance, MBF and pressure can be obtained.38 CFD itself is not a new concept and was used to solve equations pertaining to blood flow for the past 20 years.39,40 The physical properties of blood such as fluid viscosity and density are assumed when solving the equations for coronary blood flow. Since blood is an incompressible fluid, it can be treated as a Newtonian fluid with a relatively constant viscosity in larger vascular beds such as coronary arteries. The governing equations of blood flow are non-linear and can only be solved analytically for highly idealized problems. For that reason, clinical evaluation of a true vascular territory such as coronary arteries requires a numerical method to estimate the principle equations and generate a solution, which must be solved in a simultaneous fashion over many time intervals in a single cardiac cycle, using CFD methods.38 In order to solve flow equations, flow boundaries should be defined. In the case of blood flow in coronary arteries, these boundaries include the vessel lumen, the root of aorta as the inlet, and the thoracic aorta and the coronary arteries as the outlets. The model of left ventricular connection with the aorta is created at the aortic inlet.41 To derive a non-invasive computational FFR, a simplified model of the coronary and systemic circulation will be coupled with the three-dimensional model of the root of the aorta and epicardial coronary arteries obtained from coronary CTA data. To achieve accurate results, the cardiac output computed in the model needs to be correlated with the one derived from the allometric scaling law. To achieve that, it is important to ensure that the aortic pressure used in the model matches the patient's measured brachial pressure.37 Myocardial volume and mass extracted from the coronary CTA data are used to calculate the overall coronary flow under resting conditions. Since coronary flow and resistance have a consistent relationship, total coronary resistance can be calculated using the coronary flow. From the calculated total coronary resistance, using the laws of morphometry, coronary resistance in each branch vessel could be calculated depending on the branch size and diameter. Using a simulation model of the effect of adenosine on reducing the peripheral resistance of the coronary microcirculation down-stream to the epicardial coronary arteries, maximum hyperemia state is estimated in the FFRCT model. The method used in this model is identical to the one used in invasive FFR, where the hyperemic resistance in the microcirculation distal to the stenosis is assumed to be the same as the microcirculatory hyperemic resistance in the hypothetical case of epicardial coronary arteries that are free of disease.42 After the modeling of epicardial coronary arteries and ascending aorta and their integration with boundary conditions for rest and hyperemic conditions is complete, FFRCT can then be obtained solving the equations of blood flow for velocity and pressure. This will eventually result in a three-dimensional model with a complete spatial distribution of FFR allowing the

5

interrogation of the coronary arteries anywhere in the model.38 Fig 4 demonstrates a clinical example of FFRCT with invasive angiography and FFR correlation.

Clinical evidence for FFRCT Even though FFRCT technology is a relatively recent development, it has been shown to be more accurate than CTA in predicting lesion specific ischemia, as determined by invasive FFR. Table 2 summarizes the relative strengths and limitations of FFRCT. To date, 3 separate multicenter prospective studies have examined the performance of FFRCT using invasive FFR as gold standard:

DISCOVER-FLOW study 43 The prospective, multicenter international Diagnosis of ISChemiaCausing Stenoses Obtained Via Non-invasivE FRactional FLOW Reserve (DISCOVER-FLOW) trial43 was the first study to evaluate the accuracy of FFRCT in a prospective fashion. In this study, 103 patients underwent CTA and invasive FFR measurement and the per-vessel and per-patient performance of FFRCT was compared to CTA in relationship with invasive FFR as the gold standard. In the per-patient analysis, FFRCT was superior to CTA as a predictor of lesion specific ischemia with higher accuracy, specificity, positive predictive value (PPV), and negative predictive value (NPV) (Table 1). Similarly, in the per-vessel analysis, FFRCT demonstrated higher accuracy, specificity, PPV, and NPV compared to CTA for predicting lesion specific ischemia (Table 3). There was a significant correlation between FFRCT and FFR values for per patient and per vessel analysis (Pearson's correlation coefficient 0.678, p < 0.001). Interestingly, both FFRCT and FFR were not well correlated to percent diameter stenosis by quantitative coronary angiography (Pearson's correlation coefficient 0.38, and 0.43, respectively), reinforcing the NIPSS and PINSS concepts described above. Among the 47 vessels with 50–69% stenosis by CTA, 25.5% had invasive FFR-verified ischemia. Within this group FFRCT significantly increased the accuracy, sensitivity, specificity, PPV and NPV to 83.0%, 66.7%, 88.6%, 66.7%, and 88.6%, respectively. A sub-analysis of the DISCOVER-FLOW study44 examined 60 patients (58% of total sample) with a total of 66 lesions with coronary angiography verified diameter stenosis of 40–69%. In this sub analysis, FFRCT showed better performance values in all accounts compared to CTA (Table 3). The effect of image quality on the performance of FFRCT was evaluated and it was shown that FFRCT was superior to CTA in determining lesion specific ischemia, regardless of image quality.45 Specifically, in the scans with motion artifact, calcium artifact or low signal to noise ratio, FFRCT performed significantly better than CTA.

DEFACTO trial 46 Determination of Fractional Flow Reserve from Anatomic Computed Tomographic Angiography (DEFACTO) trial is a multicenter study in which 252 patients and 407 vessels were

Please cite this article as: Hulten E, et al. CT Assessment of Myocardial Perfusion and Fractional Flow Reserve. Prog Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.pcad.2015.03.003

6

P R O GRE S S I N CA RDI O VAS CU L AR D I SE AS E S XX ( 2 0 15 ) XXX– X XX

PCoronary PAorta

FFR CT 0.93

FFR CT 0.66 FFRCT 0.74

CCTA >70% LAD stenosis

QCA 54% LAD stenosis

FFRCT 0.74 FFR 0.77 Fig 4 – Example of FFRCT from a 66-year-old man with symptoms of possible angina and no known coronary disease with risk factors of diabetes, hypertension, hyperlipidemia, and prior smoking. In the mid left anterior descending coronary artery, a severe (>70%) stenosis by CTA was found to have FFRCT of 0.74 (< 0.8 significant). The same lesion was determined by quantitative invasive angiography to have 54% stenosis with an invasive FFR of 0.77. The patient also had a high grade stenosis in the distal right coronary circulation (FFRCT 0.66). Images courtesy of Dr. Bon-Kwon Koo, Seoul National University Hospital, Seoul, Korea.

investigated by both FFRCT and invasive FFR. The primary endpoint of this study was to assess the diagnostic accuracy of FFRCT compared to invasive FFR. Similar to DISCOVER-FLOW study, DEFACTO trial revealed that on a per-patient analysis, FFR CT was superior to CT stenosis ≥ 50% in predicting lesion specific ischemia (Table 3). The primary end point of the study was set as having the lower bound of confidence interval for diagnostic accuracy to be more than 70%, as that represents a 15% increase in diagnostic accuracy over stress echocardiography or MPI when compared to invasive FFR. The per-patient diagnostic accuracy for FFRCT plus CT was 73% (95% CI, 67–78%), which did not meet the pre-specified primary end point. While the performance of FFRCT did not meet the criteria for the pre-specified primary endpoint, DEFACTO confirmed the superiority of FFRCT over CT alone for the determination of lesion specific ischemia as the diagnostic accuracy of CTA

alone was 64% (95% CI 58–70%) (Table 3) and FFRCT showed 9% absolute improvement in the diagnostic accuracy. Discrimination was quantified using the area under the receiver operating characteristic curve (AUC). Discrimination of FFRCT as compared to CTA stenosis ≥ 50% was significantly higher (AUC 0.81 vs. 0.68, p = 0.0002). Importantly, there was a significant improvement seen in the evaluation of intermediate severity lesions causing 30–70% luminal narrowing.

NXT trial 47 The Analysis of Coronary Blood Flow Using CT Angiography: Next Steps (NXT) trial is the third multicenter prospective study that has evaluated the diagnostic performance of FFRCT using invasive FFR as a gold standard. This study evaluated the diagnostic accuracy of an updated iteration of FFRCT, and for the first time compared the FFRCT accuracy to detect lesion specific ischemia to both CTA and ICA.

Please cite this article as: Hulten E, et al. CT Assessment of Myocardial Perfusion and Fractional Flow Reserve. Prog Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.pcad.2015.03.003

7

PR O G RE S S I N C ARDI O V A S CU L A R D I S EA S E S XX ( 2 0 1 5) XXX – XXX

Table 2 – Strengths and limitations of FFRCT. Strengths: Simultaneous data on coronary anatomy and physiology using single CTA study Provides measure for lesion specific ischemia No additional radiation, contrast, or medications compared to routine CTA Fast and well tolerated exam Limitations: Off-site processing (results typically provided within 24 hours) Adequate CTA image quality required Further data needed on clinical effectiveness outside of clinical trials.

Two hundred fifty four patients scheduled to undergo clinically indicated ICA for suspected CAD were enrolled in this prospective multicenter trial. Patients had a coronary CTA before ICA. Diagnostic accuracy of FFR CT, CTA and ICA was compared to invasive FFR, with ischemia defined as FFR ≤ 0.80. The primary study end point was per-patient diagnostic performance as assessed by the area under the receiveroperating characteristic curve (AUC). This study compared FFRCT (≤0.80) versus coronary CTA (stenosis >50%) against the reference standard of invasive FFR ≤0.80 in patients with coronary CTA stenosis of 30–90%. Secondary end points included assessment of diagnostic accuracy, sensitivity, specificity, PPV, and NPV of FFRCT compared to coronary CTA and ICA in detecting lesion specific ischemia detected by invasive FFR. The area under the receiver-operating characteristic curve of FFRCT was 0.90 (95% CI 0.87–0.94) versus 0.81 (95% CI: 0.76–0.87) for coronary CTA (p = 0.0008). Similar to the previous studies, the per-patient and per-vessel sensitivity, specificity, PPV and NPV of FFRCT were superior to CTA for detecting lesion specific ischemia (Table 3). FFRCT was also superior in diagnostic accuracy for lesion specific ischemia compared to ICA lesion >50% (Table 3).

To date, one study has evaluated the cost effectiveness of FFRCT using data from the DISCOVER-FLOW trial.43 Hlatky et al. evaluated data from 96 patients and modeled the cost of using CTA with FFRCT versus ICA with visual assessment of stenosis.48 This study concluded that the cost of the FFRCT strategy ($7674) was 30% lower than ICA, principally due to deferred PCI. Additionally, the modeled event rate would be 12% lower annualized incidence of death or non-fatal myocardial infarction (2.63% annualized incidence decreased to 2.31%). FFRCT is a novel technology which allows physicians to obtain information about anatomy and physiology of CAD by providing details on plaque burden, location, composition and lesion specific ischemia through the integration of computational fluid dynamics and an accurate three-dimensional model of the coronary arteries. Thus far, three large multicenter prospective studies showed superiority of FFRCT over CTA in detecting lesion specific ischemia detected by FFR in both per patients and per vessel bases. Importantly, FFRCT has been shown to be superior to both ICA and CTA in detecting lesion specific ischemia in intermediate lesions; such lesions are likely the ones in which FFRCT is likely to offer the greatest benefit.

Conclusions All studies that have evaluated the diagnostic accuracy of FFRCT, have done so by comparing FFRCT with invasive FFR measurements in a binary fashion and have defined presence of lesion specific ischemia with a cut point of FFR ≤0.80. Even though using FFR ≤0.80 as binary cut point to define lesion specific ischemia is a common clinical practice, it has been recently described that FFR displays a continuous relationship between its numeric value and prognosis.49 Therefore, a study that compares the diagnostic accuracy of FFRCT with invasive FFR as a continuous variable would be a potential suitable next step in evaluating the value of FFRCT. Based upon these

Table 3 – Summary of FFRCT trials. Study DISCOVER-FLOW Per vessel (n = 159) FFRCT ≤0.8 CTA stenosis ≥ 50% Per patient (n = 103) FFRCT ≤0.8 CTA stenosis ≥ 50% DEFACTO Per patient (n = 252) FFRCT ≤0.8 CTA stenosis ≥ 50% NXT Per vessel (n = 484) FFRCT ≤0.8 CTA stenosis ≥ 50% ICA stenosis ≥ 50% Per patient (n = 251) FFRCT ≤ 0.8 CTA stenosis ≥ 50% ICA stenosis ≥ 50%

Accuracy

Sensitivity

Specificity

PPV

NPV

84 (78–90) 59 (50–66)

88 (77–95) 91 (81–97)

82 (73–89) 40 (30–50)

74 (62–84) 47 (37–56)

92 (85–97) 89 (76–96)

87 (79–93) 61 (51–71)

93 (82–98) 94 (85–99)

82 (68–91) 25(13–39)

85 (73–93) 58 (47–68)

91 (78–98) 80 (52–96)

73 (67–78) 64 (58–70)

90 (84–95) 84 (77–90)

54 (46–83) 42 (34–51)

67 (60–74) 61 (53–67)

84 (74–90) 72 (61–81)

86 (83–89) 65 (61–69) 82 (79–86)

84 (75–89) 83 (74–89) 55 (45–65)

86 (82–89) 60 (56–65) 90 (86–93)

61 (53–69) 33 (27–39) 58 (48–68)

95 (93–97) 92 (88–95) 88 (85–92)

81 (76–85) 53 (47–57) 77 (71–82)

86 (77–92) 94 (86–97) 64 (53–74)

79 (72–84) 34 (27–41) 83 (77–88)

65 (56–74) 40 (33–47) 63 (52–73)

93 (87–96) 92 (83–97) 83 (77–89)

Please cite this article as: Hulten E, et al. CT Assessment of Myocardial Perfusion and Fractional Flow Reserve. Prog Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.pcad.2015.03.003

8

P R O GRE S S I N CA RDI O VAS CU L AR D I SE AS E S XX ( 2 0 15 ) XXX– X XX

encouraging initial trials, the United States Food and Drug Administration recently approved clinical use of HeartFlow, Inc's FFRCT software. However, one must acknowledge that even though use of CT to assess coronary anatomy simultaneously with physiology is intriguing and promising, it has not yet been evaluated in real world clinical practice. There is no cost-effectiveness data to support that such methods will result in improvements in patient outcomes or cost savings when used to guide the therapy of stable CAD patients compared to the currently available non-invasive and invasive approaches.

10.

11.

12.

Statement of Conflict of Interest The authors have no financial conflicts. The opinions and assertions herein are the authors' alone and do not represent the views of the Walter Reed National Military Medical Center, the US Army, or the Department of Defense.

13.

14.

REFERENCES

1. Moshage WE, Achenbach S, Seese B, Bachmann K, Kirchgeorg M. Coronary artery stenoses: three-dimensional imaging with electrocardiographically triggered, contrast agent-enhanced, electron-beam CT. Radiology. 1995;196:707-714. 2. Mowatt G, Cook JA, Hillis GS, et al. 64-Slice computed tomography angiography in the diagnosis and assessment of coronary artery disease: systematic review and meta-analysis. Heart. 2008;94:1386-1393. 3. Voros S, Rinehart S, Qian Z, et al. Coronary atherosclerosis imaging by coronary CT angiography: current status, correlation with intravascular interrogation and meta-analysis. J Am Coll Cardiol Img. 2011;4:537-548. 4. Hulten EA, Carbonaro S, Petrillo SP, Mitchell JD, Villines TC. Prognostic value of cardiac computed tomography angiography: a systematic review and meta-analysis. J Am Coll Cardiol. 2011;57: 1237-1247. 5. Taylor AJ, Cerqueira M, Hodgson JM, et al. ACCF/SCCT/ACR/AHA/ASE/ASNC/NASCI/SCAI/SCMR 2010 Appropriate Use Criteria for Cardiac Computed Tomography. A Report of the American College of Cardiology Foundation Appropriate Use Criteria Task Force, the Society of Cardiovascular Computed Tomography, the American College of Radiology, the American Heart Association, the American Society of Echocardiography, the American Society of Nuclear Cardiology, the North American Society for Cardiovascular Imaging, the Society for Cardiovascular Angiography and Interventions, and the Society for Cardiovascular Magnetic Resonance. Circulation. 2010;122:e525-e555. 6. Ahmadi A, Kini A, Narula J. Discordance Between Ischemia and Stenosis, or PINSS and NIPSS: Are We Ready for New Vocabulary? J Am Coll Cardiol Img. 2015;8:111-114. 7. Meijboom WB, Van Mieghem CA, van Pelt N, et al. Comprehensive assessment of coronary artery stenoses: computed tomography coronary angiography versus conventional coronary angiography and correlation with fractional flow reserve in patients with stable angina. J Am Coll Cardiol. 2008;52:636-643. 8. Califf RM, Phillips III HR, Hindman MC, et al. Prognostic value of a coronary artery jeopardy score. J Am Coll Cardiol. 1985;5: 1055-1063. 9. Chow BJ, Small G, Yam Y, et al. Incremental prognostic value of cardiac computed tomography in coronary artery disease using

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

CONFIRM: COroNary computed tomography angiography evaluation for clinical outcomes: an InteRnational Multicenter registry. Circ Cardiovasc Imaging. 2011;4:463-472. Boden WE, O'Rourke RA, Teo KK, et al. Optimal medical therapy with or without PCI for stable coronary disease. N Engl J Med. 2007;356:1503-1516. Hachamovitch R, Hayes SW, Friedman JD, Cohen I, Berman DS. Comparison of the short-term survival benefit associated with revascularization compared with medical therapy in patients with no prior coronary artery disease undergoing stress myocardial perfusion single photon emission computed tomography. Circulation. 2003;107:2900-2907. Hachamovitch R, Rozanski A, Shaw LJ, et al. Impact of ischaemia and scar on the therapeutic benefit derived from myocardial revascularization vs. medical therapy among patients undergoing stress-rest myocardial perfusion scintigraphy. Eur Heart J. 2011;32:1012-1024. De Bruyne B, Pijls NH, Kalesan B, et al. Fractional flow reserve-guided PCI versus medical therapy in stable coronary disease. N Engl J Med. 2012;367:991-1001. Pijls NH, Fearon WF, Tonino PA, et al. Fractional flow reserve versus angiography for guiding percutaneous coronary intervention in patients with multivessel coronary artery disease: 2-year follow-up of the FAME (Fractional Flow Reserve Versus Angiography for Multivessel Evaluation) study. J Am Coll Cardiol. 2010;56:177-184. Tonino PA, De Bruyne B, Pijls NH, et al. Fractional flow reserve versus angiography for guiding percutaneous coronary intervention. N Engl J Med. 2009;360:213-224. Tonino PA, Fearon WF, De Bruyne B, et al. Angiographic versus functional severity of coronary artery stenoses in the FAME study fractional flow reserve versus angiography in multivessel evaluation. J Am Coll Cardiol. 2010;55:2816-2821. George RT, Jerosch-Herold M, Silva C, et al. Quantification of myocardial perfusion using dynamic 64-detector computed tomography. Investig Radiol. 2007;42:815-822. Blankstein R, Shturman LD, Rogers IS, et al. Adenosine-induced stress myocardial perfusion imaging using dual-source cardiac computed tomography. J Am Coll Cardiol. 2009;54:1072-1084. George RT, Arbab-Zadeh A, Miller JM, et al. Adenosine stress 64- and 256-row detector computed tomography angiography and perfusion imaging: a pilot study evaluating the transmural extent of perfusion abnormalities to predict atherosclerosis causing myocardial ischemia. Circ Cardiovasc Imaging. 2009;2:174-182. Okada DR, Ghoshhajra BB, Blankstein R, et al. Direct comparison of rest and adenosine stress myocardial perfusion CT with rest and stress SPECT. J Nucl Cardiol. 2010;17:27-37. Bamberg F, Becker A, Schwarz F, et al. Detection of hemodynamically significant coronary artery stenosis: incremental diagnostic value of dynamic CT-based myocardial perfusion imaging. Radiology. 2011;260:689-698. Patel AR, Lodato JA, Chandra S, et al. Detection of myocardial perfusion abnormalities using ultra-low radiation dose regadenoson stress multidetector computed tomography. J Cardiovasc Comput Tomogr. 2011;5:247-254. Tamarappoo BK, Dey D, Nakazato R, et al. Comparison of the extent and severity of myocardial perfusion defects measured by CT coronary angiography and SPECT myocardial perfusion imaging. J Am Coll Cardiol Img. 2010;3:1010-1019. Ho K-T, Chua K-C, Klotz E, Panknin C. Stress and rest dynamic myocardial perfusion imaging by evaluation of complete time-attenuation curves with dual-source CT. J Am Coll Cardiol Img. 2010;3:811-820. Feuchtner G, Goetti R, Plass A, et al. Adenosine Stress High-Pitch 128-Slice Dual-Source Myocardial Computed

Please cite this article as: Hulten E, et al. CT Assessment of Myocardial Perfusion and Fractional Flow Reserve. Prog Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.pcad.2015.03.003

PR O G RE S S I N C ARDI O V A S CU L A R D I S EA S E S XX ( 2 0 1 5) XXX – XXX

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

Tomography Perfusion for Imaging of Reversible Myocardial Ischemia Comparison With Magnetic Resonance Imaging. Circ Cardiovasc Imaging. 2011;4:540-549. Ko BS, Cameron JD, Meredith IT, et al. Computed tomography stress myocardial perfusion imaging in patients considered for revascularization: a comparison with fractional flow reserve. Eur Heart J. 2012;33:67-77. Fazel R, Gerber TC, Balter S, et al. Approaches to Enhancing Radiation Safety in Cardiovascular Imaging A Scientific Statement From the American Heart Association. Circulation. 2014;130:1730-1748. Chen J, Einstein AJ, Fazel R, et al. Cumulative exposure to ionizing radiation from diagnostic and therapeutic cardiac imaging procedures: a population-based analysis. J Am Coll Cardiol. 2010;56:702-711. Ghoshhajra BB, Rogers IS, Maurovich-Horvat P, et al. A comparison of reconstruction and viewing parameters on image quality and accuracy of stress myocardial CT perfusion. J Cardiovasc Comput Tomogr. 2011;5:459-466. Blankstein R, Rogers IS, Cury RC. Practical tips and tricks in cardiovascular computed tomography: diagnosis of myocardial infarction. J Cardiovasc Comput Tomogr. 2009;3:104-111. Rochitte CE, George RT, Chen MY, et al. Computed tomography angiography and perfusion to assess coronary artery stenosis causing perfusion defects by single photon emission computed tomography: the CORE320 study. Eur Heart J. 2014;35:1120-1130. George RT, Mehra VC, Chen MY, et al. Myocardial CT perfusion imaging and SPECT for the diagnosis of coronary artery disease: a head-to-head comparison from the CORE320 multicenter diagnostic performance study. Radiology. 2014;272:407-416. Cury RC, Kitt TM, Feaheny K, Akin J, George RT. Regadenoson-stress myocardial CT perfusion and single-photon emission CT: rationale, design, and acquisition methods of a prospective, multicenter, multivendor comparison. J Cardiovasc Comput Tomogr. 2014;8:2-12. Cury RC, Kitt TM, Feaheny K, et al. A Randomized, Multicenter, Multivendor Study of Myocardial Perfusion Imaging With Regadenoson Computed Tomography Perfusion versus Single Photon Emission Computed Tomography. J Cardiovasc Comput Tomogr. 2015. Weininger M, Schoepf UJ, Ramachandra A, et al. Adenosine-stress dynamic real-time myocardial perfusion CT and adenosine-stress first-pass dual-energy myocardial perfusion CT for the assessment of acute chest pain: Initial results. European journal of radiology;81:3703–3710. Truong QA, Knaapen P, Pontone G, et al. Rationale and design of the dual-energy computed tomography for ischemia determination compared to "gold standard" non-invasive and invasive techniques (DECIDE-Gold): A multicenter international efficacy diagnostic study of rest-stress dual-energy computed tomography angiography with perfusion. J Nucl Cardiol. 2014.

9

37. Kim HJ, Vignon-Clementel IE, Coogan JS, Figueroa CA, Jansen KE, Taylor CA. Patient-specific modeling of blood flow and pressure in human coronary arteries. Ann Biomed Eng. 2010;38: 3195-3209. 38. Grunau GL, Min JK, Leipsic J. Modeling of fractional flow reserve based on coronary CT angiography. Curr Cardiol Rep. 2013;15:336. 39. Perktold K, Resch M, Peter RO. Three-dimensional numerical analysis of pulsatile flow and wall shear stress in the carotid artery bifurcation. J Biomech. 1991;24:409-420. 40. Taylor CA, Hughes TJ, Zarins CK. Finite element modeling of three-dimensional pulsatile flow in the abdominal aorta: relevance to atherosclerosis. Ann Biomed Eng. 1998;26:975-987. 41. Kim HJ, Vignon-Clementel IE, Figueroa CA, et al. On coupling a lumped parameter heart model and a three-dimensional finite element aorta model. Ann Biomed Eng. 2009;37: 2153-2169. 42. Pijls NH, van Son JA, Kirkeeide RL, De Bruyne B, Gould KL. Experimental basis of determining maximum coronary, myocardial, and collateral blood flow by pressure measurements for assessing functional stenosis severity before and after percutaneous transluminal coronary angioplasty. Circulation. 1993;87:1354-1367. 43. Koo BK, Erglis A, Doh JH, et al. Diagnosis of ischemia-causing coronary stenoses by noninvasive fractional flow reserve computed from coronary computed tomographic angiograms. Results from the prospective multicenter DISCOVER-FLOW (Diagnosis of Ischemia-Causing Stenoses Obtained Via Noninvasive Fractional Flow Reserve) study. J Am Coll Cardiol. 2011;58:1989-1997. 44. Min JK, Koo BK, Erglis A, et al. Usefulness of noninvasive fractional flow reserve computed from coronary computed tomographic angiograms for intermediate stenoses confirmed by quantitative coronary angiography. Am J Cardiol. 2012;110: 971-976. 45. Min JK, Koo BK, Erglis A, et al. Effect of image quality on diagnostic accuracy of noninvasive fractional flow reserve: results from the prospective multicenter international DISCOVER-FLOW study. J Cardiovasc Comput Tomogr. 2012;6:191-199. 46. Min JK, Leipsic J, Pencina MJ, et al. Diagnostic accuracy of fractional flow reserve from anatomic CT angiography. JAMA. 2012;308:1237-1245. 47. Norgaard BL, Leipsic J, Gaur S, et al. Diagnostic performance of noninvasive fractional flow reserve derived from coronary computed tomography angiography in suspected coronary artery disease: the NXT trial (Analysis of Coronary Blood Flow Using CT Angiography: Next Steps). J Am Coll Cardiol. 2014;63:1145-1155. 48. Hlatky MA, Saxena A, Koo BK, Erglis A, Zarins CK, Min JK. Projected costs and consequences of computed tomography-determined fractional flow reserve. Clin Cardiol. 2013;36:743-748. 49. Johnson NP, Toth GG, Lai D, et al. Prognostic value of fractional flow reserve: linking physiologic severity to clinical outcomes. J Am Coll Cardiol. 2014;64:1641-1654.

Please cite this article as: Hulten E, et al. CT Assessment of Myocardial Perfusion and Fractional Flow Reserve. Prog Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.pcad.2015.03.003