Assessment of Functional Severity of Coronary Lesions by CT Angiography — FFRCT

Abstract & Commentary

By Andrew J. Boyle, MBBS, PhD, Assistant Professor of Medicine, Interventional Cardiology, University of California, San Francisco

Source: Koo BK, et al. Diagnosis of ischemia-causing coronary stenoses by noninvasive fractional flow reserve computed from coronary computed tomographic angiograms. J Am Coll Cardiol 2011;58:1989-1997.

Anatomic definition of coronary artery lesions by invasive coronary angiography (ICA) or computed tomography coronary angiography (CCTA) has limitations. The addition of physiologic assessment of coronary lesion severity has been shown to improve clinical decisionmaking. Fractional flow reserve (FFR) has been shown in the FAME study to safely and cost-effectively guide the decision on which lesion(s) to revascularize.1 This involves placing a pressure sensing wire within the coronary artery and derives a ratio of pressure distal to the lesion divided by pressure proximal to the lesion. An FFR ≤ 0.80 indicates that the lesion causes ischemia and should be revascularized. Lesions with FFR > 0.80 can safely be treated medically. However, this is invasive and expensive. A non-invasive test that can define the anatomy and functional significance of coronary artery lesions could improve our ability to diagnose and treat coronary artery disease (CAD). Koo and colleagues describe a novel technique to non-invasively determine the FFR from CCTA studies (FFRCT).

The authors studied 159 vessels in 103 patients that had CCTA and subsequently underwent ICA and FFR measurement. Inclusion criteria were age ≥ 18 years and a CCTA with at least one stenosis ≥ 50% in a vessel ≥ 2 mm that were undergoing clinically indicated ICA and FFR. Demographics were fairly standard for a CAD study population: mean age was 63 years, 74% were male, 65% had hypertension and dyslipidemia, and 36% were diabetic. However, mean BMI was 26 and only 33% were Caucasian. Exclusion criteria were pregnancy, life expectancy < 2 years, renal impairment, arrhythmia, contrast allergy, inability to take nitroglycerin, beta-blocker or adenosine, prior CABG, class IV angina, and uninterpretable CCTA. CCTA was acquired using standard protocols after administration of beta-blocker for heart rate ≥ 65/min and all patients received sublingual nitroglycerin. A core lab analyzed the CCTA images. The FFRCT was analyzed by scientists at the company who developed the software. Using complex mathematical modeling (computational fluid dynamics) and making assumptions about the microcirculation based on LV mass, FFRCT was calculated. It took an average of 5 hours to derive the FFRCT. Invasive FFR was performed according to standard protocols with either IV or intracoronary adenosine to achieve maximal hyperemia, according to the operator's preference.

FFRCT correlated well with invasive FFR (r = 0.717, P < 0.001) with a slight underestimation of FFRCT compared to FFR (0.02 ± 0.12, P = 0.02). Using FFR as the gold standard, FFRCT had the following sensitivity, specificity, positive-predictive value, and negative-predictive value: 87.9%, 82.2%, 73.9%, and 92.2%. Compared to using just the degree of stenosis on CCTA, FFRCT was more accurate in detecting ischemia-producing lesions. The area under the receiver operating characteristics curve, a measure of accuracy of the test, was 0.90, indicating a high level of diagnostic accuracy. The authors conclude that noninvasive FFR derived from CCTA is a novel method with high diagnostic performance for the detection and exclusion of coronary lesions that cause ischemia.


This is an interesting and important study that has potential to significantly impact the way we assess patients with chest pain. Currently, Appropriate Use Criteria advocate the use of CCTA in patients with low-to-intermediate likelihood of having CAD, whereas its use in patients with established CAD is considered uncertain or inappropriate. While the absence of CAD or the presence of mild plaque in patients with acute chest pain has been associated with low likelihood of acute coronary syndromes, the use of CCTA in lesions of moderate severity has not been determined. In this study, the minority of moderate lesions were actually functionally significant by FFR or by FFRCT. One could imagine foregoing ICA in patients with an intermediate stenosis on CCTA that has a non-significant FFRCT, although this strategy remains to be tested. Now for the first time, CCTA appears to be able to incorporate both the anatomical definition of the lesion with the physiological significance of that lesion — this is an important step forward.

Several features about this study should be noted. First, invasive FFR actually induces maximal hyperemia in the individual patient, whereas the computational fluid dynamic analysis that derives the FFRCT makes assumptions about the microvascular bed, rather than inducing hyperemia. I worry that inter-individual differences in microvascular function may lead to inaccurate diagnoses sometimes. Second, only patients undergoing ICA for ≥ 50% stenosis were included. This is important ethically in the early stage investigation of new technologies, but the exclusion of lesser degrees of stenosis may underestimate the false-positive rate. Third, the exclusion of patients with prior CABG means the performance of this modality in these patients is unknown. Finally, the huge computing power needed to derive the FFRCT means the studies took 5 hours to analyze. Future iterations of the technology are required to make it more widely applicable. Despite these limitations, this technique has significant potential. The combination of anatomy and function in a non-invasive study could improve our diagnosis and management of patients in a safer and more cost-effective way.


1. Tonino PA, et al. Fractional flow reserve versus angiography for guiding percutaneous coronary intervention. N Engl J Med 2009; 360:213-224.