Non-Invasive Computed Fractional Flow Reserve From Computed Tomography (CT) for Diagnosing Coronary Artery Disease

– Japanese Results From NXT Trial (Analysis of Coronary Blood Flow Using CT Angiography: Next Steps) –

Abstract

Background: Recently, a non-invasive method using computational fluid dynamics to calculate vessel-specific fractional flow reserve (FFR_CT) from routinely acquired coronary computed tomography angiography (CTA) was described. The Analysis of Coronary Blood Flow Using CT Angiography: Next Steps (NXT) trial, which was a prospective, multicenter trial including 254 patients with suspected coronary artery disease, noted high diagnostic performance of FFR_CT compared with invasive FFR. The aim of this post-hoc analysis was to assess the diagnostic performance of non-invasive FFR_CT vs. standard stenosis quantification on coronary CTA in the Japanese subset of the NXT trial.

Methods and Results: A total of 57 Japanese participants were included from Okayama University (n=36), Kyoto University (n=17), and Keio University (n=4) Hospitals. Per-patient diagnostic accuracy of FFR_CT (74%; 95% confidence interval [CI]: 60–85%) was higher than for coronary CTA (47%; 95% CI: 34–61%, P<0.001) arising from improved specificity (63% vs. 27%, P<0.001). FFR_CT correctly reclassified 53% of patients and 63% of vessels with coronary CTA false positives as true negatives. When patients with Agatston score >1,000 were excluded, per-patient accuracy of FFR_CT was 83% with a high specificity of 76%, similar to the overall NXT trial findings.

Conclusions: FFR_CT has high diagnostic performance compared with invasive FFR in the Japanese subset of patients in the NXT trial. (Circ J 2015; 79: 406–412)

Coronary computed tomography angiography (CTA) has diagnostic accuracy in the detection and exclusion of coronary artery disease (CAD).^1–6 Thus, coronary CTA is being increasingly applied in the work-up of patients with suspected CAD in Japan. According to the Japanese Registry Of All cardiac and vascular Diseases (JROAD) the number of coronary CTA examinations has increased more than 2-fold during the last 5 years, approaching 400,000 in 2012.⁷ Previous studies, however, have shown that stenosis severity as evaluated on coronary CTA does not correlate well with functional assessment using invasive fractional flow reserve (FFR).^8,9 Measurement of FFR using invasive coronary angiography (ICA) is the gold standard for the diagnosis of stenosis causing lesion-specific ischemia.¹⁰ The Fractional Flow Reserve vs. Angiography for Multivessel Evaluation (FAME) study has shown that invasive FFR-guided decision-making regarding coronary revascularization improves event-free survival compared with coronary angiography-guided decisions alone.¹¹

Editorial p 300

Recently, a method using computational fluid dynamics to non-invasively calculate FFR from routinely acquired coronary CTA without any additional image acquisition and medications (FFR_CT), was described.^12–16 The Analysis of Coronary Blood Flow Using CT Angiography: Next Steps (NXT) trial is the largest, prospective multicenter trial to evaluate FFR_CT, and it included 254 patients with suspected CAD from 10 sites in Europe, North America, and the Asian Pacific area. The study found high diagnostic performance of FFR_CT compared with invasively measured FFR, with accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for FFR_CT on a per-patient basis of 81%, 86%, 79%, 65%, and 93%, respectively.¹⁶

The body mass index in Japanese patients is lower than in European or North American patients, which may result in the higher image quality of coronary CTA. In contrast, Japanese patients have smaller coronary arteries, which may lower image quality of coronary CTA. Thus, the results of the overall NXT trial may be different from the Japanese cohort. The aim of this post-hoc analysis based on the NXT trial was therefore to assess the diagnostic performance of non-invasive FFR_CT compared with coronary CTA using invasive FFR as the reference standard in the Japanese subset of subjects in the NXT trial.

Methods

Study Design and Subjects

This post-hoc analysis was conducted as a substudy of the NXT trial, which has been described in detail elsewhere.^16,17 In summary, the NXT trial was a prospective, international, multicenter trial that included 254 patients scheduled to undergo clinically indicated ICA for suspected CAD. The principal aim of the study was to compare the diagnostic performance of FFR_CT with coronary CTA using invasive FFR as the reference standard. A total of 57 Japanese participants in the NXT trial were included from Okayama University (n=36), Kyoto University (n=17), and Keio University (n=4) Hospitals. Coronary CTA performed <60 days before scheduled ICA was required for inclusion. Exclusion criteria included previous coronary intervention or coronary bypass surgery; contraindications to β-blocking agents, nitroglycerin, or adenosine; suspected acute coronary syndrome; previous myocardial infarction <30 days before coronary CTA or between coronary CTA and ICA; and body mass index >35 kg/m².

This study was approved by the Ethics Committee of Okayama University Graduate School of Medicine, Dentistry, and Pharmaceutical Sciences (Okayama, Japan). This study was conducted according to the principles expressed in the Declaration of Helsinki. All participants provided written informed consent before screening and study enrollment.

Image Acquisition and Coronary CTA

Coronary CTA was performed in accordance with Society of Cardiovascular Computed Tomography guidelines.¹⁸ Oral and/or i.v. β-blockers were used to achieve a heart rate <60 beats/min, and sublingual nitrates were used for dilating coronary arteries. Assessment of luminal diameter stenosis was performed at the discretion of the local investigator using an 18-segment coronary model. Significant coronary obstruction was defined as stenosis severity >50% in a major coronary artery segment ≥2 mm in diameter. Coronary CTA was transmitted to the FFR_CT core laboratory (HeartFlow, Redwood City, CA, USA) for computational analysis.

ICA and FFR Measurement

ICA was performed according to a standard protocol.¹⁹ Measurement of FFR (PressureWire, St. Jude Medical, St. Paul, MN, USA) was performed during ICA in at least 1 vessel segment with an estimated luminal diameter ≥2 mm.¹⁷ Hyperemia was induced by i.v. adenosine at an infusion rate of 140–180 μg·kg^–1·min^–1. FFR was calculated by dividing the mean distal coronary pressure by the mean aortic pressure during hyperemia. Measured FFR ≤0.80 in a vessel ≥2 mm in diameter was considered to be diagnostic of ischemia.¹¹ Invasive coronary angiograms were transferred to the angiography/FFR core laboratory (Harrington Heart and Vascular Institute, University Hospitals, Cleveland, OH, USA) for blinded quantitative coronary angiography of all vessels.

Integration of Coronary CTA and FFR Data

FFR_CT was compared to FFR at the location of the wire pressure transducer at the time of FFR measurement. The blinded angiography/FFR core laboratory identified the location of the invasive FFR measurement and indicated the corresponding point on the FFR_CT computational model.

Computation of FFR_CT

Computation of FFR_CT was performed in a blinded manner by the FFR_CT core laboratory.¹⁷ FFR_CT was calculated using computational fluid dynamic modeling after semi-automated segmentation of the coronary arteries and left ventricular mass.¹³ Coronary blood flow and pressure were calculated under conditions simulating maximal hyperemia. FFR_CT was obtained by dividing the mean pressure distal to the coronary stenosis by the mean aortic pressure. The results provide a complete spatial distribution of FFR_CT in the coronary arteries (Figure 1).

Figure 1.

Representative Japanese subject from the NXT Study: a 65-year-old Japanese man with abnormal electrocardiogram. He had insulin-dependent type II diabetes, hyperlipidemia, former smoking (30 years). Agatston score was 537. Coronary computed tomography angiography (CTA) showed severe calcification of the left main and proximal left anterior descending artery (LAD) and >50% stenosis of the right coronary artery (RCA) and left circumflex artery (LCx) second obtuse marginal branch (OM₂). On fractional flow reserve from routinely acquired coronary CTA (FFR_CT) analysis, the identified stenoses were not functionally significant, with RCA FFR_CT=0.88 and LCx OM₂ FFR_CT=0.95, respectively. FFR measurements during coronary angiography confirmed that the lesions were not ischemia producing, with FFR=0.93 in the RCA and FFR=0.99 in the LCx OM₂. The computational model is rotated to the same projection as the angiogram to show FFR_CT at the location corresponding to the site of FFR measurement. FFR_CT is computed for all points in the coronary tree and is represented by the color scale.

Statistical Analysis

The primary endpoint of this substudy was diagnostic performance as assessed on accuracy, sensitivity, specificity, PPV, and NPV of FFR_CT vs. coronary CTA using invasive FFR as the reference standard on both a per-patient and per-vessel level. Diagnostic accuracy, sensitivity, specificity, PPV, and NPV were calculated as simple proportions with 95% confidence intervals (95% CI). Patient-based comparison of diagnostic performance characteristics (accuracy, sensitivity, specificity, PPV, and NPV) was performed using the McNemar test for paired samples or percentile bootstrap with 100,000 resamples as appropriate. To account for potential correlation between multiple vessels in the same subject, the generalized estimating equation method with an exchangeable correlation structure was used to compare paired samples in per-vessel analysis. Statistical analysis was performed using SAS version 9.3 (SAS Institute, Cary, NC, USA).

Results

Patient Characteristics

Among 76 Japanese patients who underwent study screening between September 2012 and August 2013, 9 patients were excluded by the CT core laboratory due to non-evaluable coronary CTA. Two patients were excluded because FFR measurement was performed in vessels with a diameter <2.0 mm. Eight patients were not eligible for inclusion for other reasons. Thus, 57 Japanese patients were included in the analyses (22% of the total NXT group). Baseline demographic and clinical subject characteristics are listed in Table 1. The Japanese subset had symptoms of typical angina in 73%, angina symptoms within the past 30 days in 53%, and atypical angina symptoms in 23%. The average Diamond-Forrester score²⁰ was 55% and intermediate pre-test risk was present in 53%. Compared to the overall NXT study, the Japanese subset had more coronary artery calcification, with a mean Agatston score of 593 compared to 302 for NXT. The Agatston score was >400 in 45% of the Japanese patients compared to 26% in NXT. Nonetheless, fewer Japanese patients had lesion-specific ischemia with FFR_CT ≤0.80 in ≥1 vessels in 28% as compared to 42% of patients in NXT. Beta-blockers were used in 90% and nitrates were given in 100% of Japanese patients at the time of the coronary CTA. Coronary CTA acquisition was predominantly retrospective (98% of scans, vs. 46% in NXT) and the median radiation dose of CT was 20 mSv (IQR, 17–23 mSv). Mean time interval between coronary CTA and ICA was 16 days (range, 1–53 days).

Table 1. Subject Baseline Characteristics

	NXT trial (n=254)	Japan cohort (n=57)
Age (years)	64±10	69±9
Male gender	162 (64)	36 (63)
BMI (kg/m2)	26±3	23±3
Diabetes mellitus	58 (23)	34 (60)
Hypertension	174 (69)	43 (75)
Dyslipidemia	200 (79)	46 (81)
Current smoker	46 (18)	8 (14)
Prior MI	5 (2)	0 (0)
Symptoms
Typical angina	103 (52)	22 (73)
Atypical angina	74 (37)	7 (23)
Angina within the past month	198 (78)	30 (53)
Dyspnea	12 (6)	1 (3)
Updated Diamond-Forrester risk score (%)	58	55
Intermediate (20–80%) pre-test risk	220 (87)	30 (53)
Coronary CTA characteristics
Agatston score (mean)	302	593
Agatston score >400	66 (26)	26 (45)
Coronary CTA stenosis >50% in ≥1 vessel	221 (87)	53 (93)
Patients with FFRCT ≤0.80	106 (42)	16 (28)
Vessels with FFRCT ≤0.80	135 (28)	11 (20)
Coronary CTA acquisition
Pre-scan nitrates	253 (100)	57 (100)
Pre-scan β-blockers	198 (78)	51 (90)
Heart rate (beats/min)	63 (37–110)	68 (46–100)
Prospective acquisition	173 (54)	1 (2)
Retrospective acquisition	117 (46)	56 (98)

Data given as mean±SD, n (%) or mean (range). BMI, body mass index; CTA, computed tomography angiography; FFR_CT, fractional flow reserve derived from standard acquired coronary CTA datasets; MI, myocardial infarction; NXT, Next Steps.

FFR_CT and Coronary CTA for Diagnosis of Ischemia

A representative example of a diabetic Japanese patient with an Agatston score of 537 and >50% stenoses in 2 coronary arteries is shown in Figure 1. Table 2 compares diagnostic performance of FFR_CT and coronary CTA. Diagnostic accuracy, sensitivity, specificity, PPV and NPV for FFR_CT on a per-patient basis were 74%, 100%, 63%, 52%, and 100%, respectively. On a per-vessel basis, they were 78%, 100%, 73%, 49%, and 100%. FFR_CT correctly reclassified 53% of patients with coronary CTA false-positive findings and 63% of those with coronary CTA false-positive vessels as true-negative findings (Figure 2).

Table 2. FFR_CT vs. Coronary CTA in Japanese NXT Subset (n=57)

	Per patient			Per vessel
	FFRCT ≤0.80	Coronary CTA stenosis >50%	P-value	FFRCT ≤0.80	Coronary CTA stenosis >50%	P-value
	% (95% CI)	% (95% CI)		% (95% CI)	% (95% CI)
Accuracy	74.0 (60.3–84.5)	47.0 (34.0–61.0)	<0.001	78.0 (68.4–86.5)	52.3 (41.4–63.0)	<0.001
Sensitivity	100.0 (79.4–100.0)	100.0 (79.4–100.0)	NS	100.0 (81.5–100.0)	88.9 (65.3–98.6)	NS
Specificity	63.4 (46.9–77.9)	26.8 (14.2–42.9)	0.001	72.9 (60.9–82.8)	42.9 (31.1–55.3)	0.001
PPV	51.6 (33.1–69.8)	34.8 (21.4–50.2)	0.07	48.6 (31.9–65.6)	28.6 (17.3–42.2)	0.03
NPV	100.0 (86.8–100.0)	100.0 (71.5–100.0)	NS	100.0 (93.0–100.0)	93.8 (79.2–99.2)	NS

CI, confidence interval; NPV, negative predictive value; PPV, positive predictive value. Other abbreviations as in Table 1.

Figure 2.

Agreement for detection of ischemia (fractional flow reserve [FFR] ≤0.80) between (A) coronary computed tomography angiography (CTA) and (B) FFR from routinely acquired coronary CTA (FFR_CT) on a per-patient and per-vessel basis. Data given as % (95% confidence interval).

FFR_CT and Coronary CTA in Patients With Agatston Score ≤1,000

Table 3 compares diagnostic performance of FFR_CT and coronary CTA in patients with Agatston score ≤1,000 (n=47). Per-patient diagnostic accuracy of FFR_CT improved to 83% (95% CI: 69–92%) and specificity improved to 76% (95% CI: 58–89%). Similarly, per-vessel accuracy of FFR_CT improved to 85% (95% CI: 75–92%) with a specificity of 81% (95% CI: 69–90%). There were no differences in sensitivity, PPV or NPV. Diagnostic performance of FFR_CT was significantly improved compared to coronary CTA alone.

Table 3. FFR_CT vs. Coronary CTA in Japanese Patients With Agatston Score ≤1,000 (n=47)

	Per patient			Per vessel
	FFRCT ≤0.80	Coronary CTA stenosis >50%	P-value	FFRCT ≤0.80	Coronary CTA stenosis >50%	P-value
	% (95% CI)	% (95% CI)		% (95% CI)	% (95% CI)
Accuracy	83 (69–92.4)	51.1 (36.1–65.9)	<0.01	85.1 (75.0–92.3)	55.4 (43.4–67.0)	<0.01
Sensitivity	100.0 (76.8–100.0)	100.0 (76.8–100.0)	NS	100.0 (79.4–100.0)	87.5 (61.7–98.4)	NS
Specificity	75.8 (57.7–88.9)	30.3 (15.6–48.7)	<0.01	81.0 (68.6–90.1)	46.6 (33.3–60.1)	<0.01
PPV	63.6 (40.7–50.2)	37.8 (22.5–55.2)	<0.02	59.3 (38.8–77.6)	31.1 (18.2–46.6)	<0.02
NPV	100.0 (86.3–100)	100.0 (69.2–100.0)	NS	100.0 (92.5–100.0)	93.1 (77.2–99.2)	NS

Abbreviations as in Tables 1,2.

Discussion

This study demonstrates that non-invasive determination of FFR_CT improves the diagnostic accuracy compared to coronary CTA alone for the diagnosis of ischemia in the Japanese subset of the NXT trial. This improvement is primarily due to a marked improvement in specificity, from 27% for CTA to 63% for FFR_CT. FFR_CT correctly reclassified 53% of patients and 63% of vessels with false-positive CTA findings to true negatives. FFR_CT is a novel method that applies computational fluid dynamics to derive physiologic data from standard coronary CTA obtained under resting conditions with no additional radiation, contrast, or medication. Of note, the patients included in this substudy had higher coronary calcium scores than the overall NXT study cohort.¹ In a subanalysis excluding patients with Agatston score >1,000, the diagnostic accuracy and specificity of FFR_CT was similar to that in the main NXT study. Thus, the high sensitivity and NPV in this study convincingly support the usefulness of FFR_CT for the diagnosis of lesion-specific ischemia in Japanese patients with suspected CAD.

The present results are supported by previous studies of FFR_CT. The pilot Diagnosis of Ischemia-Causing Stenoses Obtained Via Non-invasive Fractional Flow Reserve (DISCOVER-FLOW) trial included 103 subjects and noted a per-vessel sensitivity of 88% and a specificity of 82% compared with invasively measured FFR.¹² The Determination of Fractional Flow Reserve by Anatomic Computed Tomographic Angiography (DeFACTO) trial enrolled 252 patients and noted a per-patient sensitivity of 90% but a specificity of only 54%, which did not meet the pre-specified primary endpoint of a lower 95% CI border for diagnostic accuracy of >70%.¹⁴ In contrast to these studies, in the NXT trial coronary CTA was acquired according to Society guidelines and FFR_CT analysis utilized the newest software with automated image processing and physiological models of microcirculatory resistance.¹⁶ Therefore, in the NXT trial most patients received β-blockers and sublingual nitrates before coronary CTA. In the Japanese substudy, 90% of participants received β-blockers and 100% received nitrates prior to coronary CTA. As a result, the specificity was 100% and NPV, 100%. Although this substudy did not use receiver operating characteristic curves as reported in the NXT trial, due to the limited number of patients, these results suggest that FFR_CT has considerable discriminatory power to identify and exclude ischemia in Japanese patients with suspected CAD.

In the NXT trial the diagnostic accuracy, sensitivity, specificity, PPV, and NPV for FFR_CT on a per-patient basis were 81%, 86%, 79%, 65%, and 93%, respectively.¹⁶ Compared to the primary study, this substudy found a lower diagnostic accuracy for FFR_CT. It is well known, however, that image quality is influenced by several conditions including heart rate and the presence of coronary artery calcium.¹ One possible reason for the difference in results between this substudy and the main NXT trial is that 45% of the patients in the Japanese subset had Agatston score >400, compared to only 26% in the NXT trial. Despite the high level of coronary calcification, in the Japanese subset of patients the sensitivity and NPV for FFR_CT were both 100%. Recently, Norgaard et al reported that the per-patient diagnostic accuracy of FFT_CT in the NXT trial was >80% in patients with Agatston score ≤1,000, and was considerably improved over coronary CTA and ICA as Agatston score increased up to 1,000 (Nørgaard BL, presented at SCCT 2014). In the present Japanese cohort, the diagnostic accuracy in patients with Agatston score <1,000 was similar to that in the overall NXT trial, suggesting that this method could apply to Japan as well.

The absence of coronary stenosis on coronary CTA is associated with an excellent clinical outcome.²¹ Coronary CTA, however, has been shown to overestimate the severity of CAD, hence the diagnostic specificity is only modest.²² Moreover, similar to ICA, coronary CTA has only a modest capacity to determine the hemodynamic significance of a stenosis.^8,23 FFR_CT extends the potential of coronary CTA to not only exclude or detect CAD, but also to decrease the number of unnecessary ICA. Moreover, FFR_CT has the advantage of non-invasively pinpointing the specific coronary lesions causing the symptoms. In accordance with the present substudy, the diagnostic performance of FFR_CT is high and superior to anatomic interpretation with coronary CTA also in patients with a high level of coronary calcification (eg, Agatston score 400–1,000).^16,24,25 Therefore, it is possible that further refinements in the CT technology and FFR_CT analysis process may reduce the impact of coronary calcification and thus in the future the clinical applicability of coronary CTA may be extended to patients at higher pre-test risk of CAD. In addition, the combination of coronary CTA and FFR_CT may be useful to decide the strategy of percutaneous coronary intervention.²⁶ Kim et al reported that virtual coronary stenting based on CT-derived computational models can predict FFR_CT after stenting.²⁷ In addition, this is an era of major concern regarding increasing health-care costs, and economic simulation analysis, based on historical data, indicates that the use of FFR_CT to guide selection of ICA and revascularization may reduce cost in patients with suspected CAD in the USA.²⁸ A similar analysis using data from the Japanese health-care system suggests that utilization of FFR_CT in Japan may result in fewer diagnostic catheterizations, fewer inappropriate percutaneous coronary interventions, improved patient outcome, and a 30% reduction in average cost per patient relative to standard care if fully implemented.²⁹ The ongoing multicenter Prospective Longitudinal Trial of FFR_CT: Outcome and Resource Impacts (PLATFORM NCT01943903) trial is prospectively exploring the benefit of FFR_CT-guided diagnostic evaluation vs. standard practice on clinical outcome and cost in patients with suspected CAD.

Study Limitations

First, the number of patients included in this substudy was relatively small, with no systemic validation cohort. Future studies on the diagnostic performance of FFR_CT involving larger Japanese cohorts are warranted. Second, given that FFR_CT is derived from coronary CTA data, significant CT artifacts such as motion, low contrast or blooming from coronary calcification may impair the diagnostic performance of FFR_CT due to the inability to accurately discern the lumen boundary. These issues can be minimized by adhering to coronary CTA acquisition guidelines, particularly by use of heart-rate lowering medication and sublingual nitrates before image acquisition. Because FFR_CT computation is based on global coronary and myocardial information, this modality may be less susceptible to single artifacts than coronary CTA per se. Accordingly, it has been shown that even at lower coronary CTA image quality, FFR_CT continues to provide significant improvement when compared with coronary CTA.²⁴ Moreover, it is anticipated that further refinements in CT technology and FFR_CT analysis may reduce the impact of CT artifacts. Third, patients with acute coronary syndrome and previous coronary intervention or bypass surgery were excluded from the present study. Thus, generalizability of FFR_CT to these specific patients with CAD is unknown. Finally, the current system requires upload of CT dicom data sets for off-site computational analysis. This is very computationally demanding and there currently is no point-of-care system available for such analysis. Reduced order models for FFR_CT computation may allow for onsite FFR_CT assessment. Further investigation in prospective multicenter trials, however, is needed in order to determine the diagnostic performance of these techniques.

Conclusions

FFR_CT has high diagnostic performance compared with invasively measured FFR in the Japanese subset of the NXT trial. Compared with coronary CTA assessment of stenosis, FFR_CT results in a marked increase in diagnostic accuracy and specificity for detection of lesion-specific ischemia. The combination of FFR_CT and coronary CTA may be useful for comprehensive non-invasive anatomic and functional assessment of CAD.

Acknowledgments

All authors have no conflict of interest.

Disclosures

This study was supported by HeartFlow, Inc.

References

• 1.    Budoff  MJ,  Dowe  D,  Jollis  JG,  Gitter  M,  Sutherland  J,  Halamert  E, et al. Diagnostic performance of 64-multidetector row coronary computed tomographic angiography for evaluation of coronary artery stenosis in individuals without known coronary artery disease: Results from the prospective multicenter ACCURACY (Assessment by Coronary Computed Tomographic Angiography of Individuals Undergoing Invasive Coronary Angiography) trial. J Am Coll Cardiol 2008; 52: 1724–1732.
• 2.    Meijboom  WB,  Meijs  MF,  Schuijf  JD,  Cramer  MJ,  Mollet  NR,  van Mieghem  CA, et al. Diagnostic accuracy of 64-slice computed tomography coronary angiography: A prospective, multicenter, multivendor study. J Am Coll Cardiol 2008; 52: 2135–2144.
• 3.    Miller  JM,  Rochitte  CE,  Dewey  M,  Arbab-Zadeh  A,  Niinuma  H,  Gottlieb  I, et al. Diagnostic performance of coronary angiography by 64-row CT. N Engl J Med 2008; 359: 2324–2336.
• 4.    Osawa  K,  Miyoshi  T,  Sato  S,  Akagi  N,  Morimitsu  Y,  Nakamura  K, et al. Safety and efficacy of a bolus injection of landiolol hydrochloride as a premedication for multidetector-row computed tomography coronary angiography. Circ J 2013; 77: 146–152.
• 5.    Urabe  Y,  Yamamoto  H,  Kitagawa  T,  Utsunomiya  H,  Tsushima  H,  Tatsugami  F, et al. Association between serum levels of n-3 polyunsaturated fatty acids and coronary plaque detected by coronary computed tomography angiography in patients receiving statin therapy. Circ J 2013; 77: 2578–2585.
• 6.    Motoyama  S,  Sarai  M,  Inoue  K,  Kawai  H,  Ito  H,  Harigaya  H, et al. Morphologic and functional assessment of coronary artery disease: Potential application of computed tomography angiography and myocardial perfusion imaging. Circ J 2013; 77: 411–417.
• 7.   Japanese Circulation Society. The Japanese Registry of All cardiac and vascular Diseases (JROAD) 2012 (in Japanese). http://jroadinfo.ncvc.go.jp/?page_id=13 (accessed January 17, 2014).
• 8.    Meijboom  WB,  Van Mieghem  CA,  van Pelt  N,  Weustink  A,  Pugliese  F,  Mollet  NR, 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.
• 9.    Schuijf  JD,  Bax  JJ. CT angiography: An alternative to nuclear perfusion imaging? Heart 2008; 94: 255–257.
• 10.    Kim  YH,  Park  SJ. Ischemia-guided percutaneous coronary intervention for patients with stable coronary artery disease. Circ J 2013; 77: 1967–1974.
• 11.    Tonino  PA,  De Bruyne  B,  Pijls  NH,  Siebert  U,  Ikeno  F,  van’t Veer  M, et al. Fractional flow reserve versus angiography for guiding percutaneous coronary intervention. N Engl J Med 2009; 360: 213–224.
• 12.    Koo  BK,  Erglis  A,  Doh  JH,  Daniels  DV,  Jegere  S,  Kim  HS, 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.
• 13.    Taylor  CA,  Fonte  TA,  Min  JK. Computational fluid dynamics applied to cardiac computed tomography for noninvasive quantification of fractional flow reserve: Scientific basis. J Am Coll Cardiol 2013; 61: 2233–2241.
• 14.    Min  JK,  Berman  DS,  Budoff  MJ,  Jaffer  FA,  Leipsic  J,  Leon  MB, et al. Rationale and design of the DeFACTO (Determination of Fractional Flow Reserve by Anatomic Computed Tomographic AngiOgraphy) study. J Cardiovasc Comput Tomogr 2011; 5: 301–309.
• 15.    Min  JK,  Leipsic  J,  Pencina  MJ,  Berman  DS,  Koo  BK,  van Mieghem  C, et al. Diagnostic accuracy of fractional flow reserve from anatomic CT angiography. JAMA 2012; 308: 1237–1245.
• 16.    Norgaard  BL,  Leipsic  J,  Gaur  S,  Seneviratne  S,  Ko  BS,  Ito  H, 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 J Am Coll Cardiol 2014; 63: 1145–1155.
• 17.    Gaur  S,  Achenbach  S,  Leipsic  J,  Mauri  L,  Bezerra  HG,  Jensen  JM, et al. Rationale and design of the HeartFlowNXT (Heartflow Analysis of Coronary Blood Flow Using CT Angiography: NeXt sTeps) study. J Cardiovasc Comput Tomogr 2013; 7: 279–288.
• 18.    Abbara  S,  Arbab-Zadeh  A,  Callister  TQ,  Desai  MY,  Mamuya  W,  Thomson  L, et al. SCCT guidelines for performance of coronary computed tomographic angiography: A report of the Society of Cardiovascular Computed Tomography Guidelines Committee. J Cardiovasc Comput Tomogr 2009; 3: 190–204.
• 19.    Naidu  SS,  Rao  SV,  Blankenship  J,  Cavendish  JJ,  Farah  T,  Moussa  I, et al. Clinical expert consensus statement on best practices in the cardiac catheterization laboratory: Society for cardiovascular angiography and interventions. Catheter Cardiovasc Interv 2012; 80: 456–464.
• 20.    Genders  TS,  Steyerberg  EW,  Alkadhi  H,  Leschka  S,  Desbiolles  L,  Nieman  K, et al. A clinical prediction rule for the diagnosis of coronary artery disease: Validation, updating, and extension. Eur Heart J 2011; 32: 1316–1330.
• 21.    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.
• 22.    Hoffmann  MH,  Shi  H,  Schmitz  BL,  Schmid  FT,  Lieberknecht  M,  Schulze  R, et al. Noninvasive coronary angiography with multislice computed tomography. JAMA 2005; 293: 2471–2478.
• 23.    Tonino  PA,  Fearon  WF,  De Bruyne  B,  Oldroyd  KG,  Leesar  MA,  Ver Lee  PN, 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.
• 24.    Min  JK,  Koo  BK,  Erglis  A,  Doh  JH,  Daniels  DV,  Jegere  S, 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.
• 25.    Leipsic  J,  Yang  TH,  Thompson  A,  Koo  BK,  Mancini  GB,  Taylor  C, et al. CT angiography (CTA) and diagnostic performance of noninvasive fractional flow reserve: Results from the Determination of Fractional Flow Reserve by Anatomic CTA (DeFACTO) study. AJR Am J Roentgenol 2014; 202: 989–994.
• 26.    Nishio  M,  Ueda  Y,  Matsuo  K,  Tsujimoto  M,  Hao  H,  Asai  M, et al. Association of target lesion characteristics evaluated by coronary computed tomography angiography and plaque debris distal embolization during percutaneous coronary intervention. Circ J 2014; 78: 2203–2208.
• 27.    Kim  KH,  Doh  JH,  Koo  BK,  Min  JK,  Erglis  A,  Yang  HM, et al. A novel noninvasive technology for treatment planning using virtual coronary stenting and computed tomography-derived computed fractional flow reserve. JACC Cardiovasc Interv 2014; 7: 72–78.
• 28.    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.
• 29.    Kimura  T,  Shiomi  H,  Kuribayashi  S,  Isshiki  T,  Kanazawa  S,  Ito  H, et al. Cost analysis of non-invasive fractional flow reserve derived from coronary computed tomographic angiography in Japan. Cardiovasc Interv Ther 2014 July 17, doi:10.1007/s12928-014-0285-1.