By Jeffrey Zimmet, MD, PhD
Associate Professor of Medicine, University of California, San Francisco; Director, Cardiac Catheterization Laboratory, San Francisco VA Medical Center
SYNOPSIS: Using a blinded, independent core laboratory, researchers compared five commercially available angiographic fractional flow reserve (FFR) methods to the pressure wire-based FFR technique. The authors found the diagnostic accuracy of these software techniques surpassed that of two-dimensional quantitative coronary angiography, but fell short of what has been reported in validation trials run by individual vendors.
SOURCE: Ninomiya K, Serruys Patrick W, Kotoku N, et al. Anonymous comparison of various angiography-derived fractional flow reserve software with pressure-derived physiological assessment. JACC Cardiovasc Interv 2023; May 10: S1936-8798(23)00760-4. doi: 10.1016/j.jcin.2023.04.026. [Online ahead of print].
Invasive coronary angiography, as it is performed in routine clinical practice, is most useful at either end of the disease severity scale. The clinical significance of intermediate-grade stenoses can be difficult to determine from two-dimensional angiography alone.
Physiologic testing using a pressure wire, a coronary guidewire with a solid-state pressure transducer, prevents most of the pitfalls of plain angiography. It is an excellent tool to use when analyzing intermediate-grade stenoses in clinical practice. Fractional flow reserve (FFR) interrogates the vessel in the setting of induced hyperemia, typically with a continuous infusion of adenosine. FFR has been replaced in many settings by instantaneous wave-free ratio (iFR) and other related nonhyperemic indices, which are faster to perform and bypass the expense and discomfort of adenosine infusion.
Despite these improvements, invasive physiologic testing is performed only in a fraction of cases where it is potentially applicable, even though clinical outcomes could improve compared to using angiography alone. A new generation of angiographic FFR indices has been developed, promising to forgo the need for a pressure wire altogether, applying a software solution to suitable angiograms. The potential to avoid the risk of coronary injury, and to assess tortuous or distal vessels that present challenges to traversing with a pressure wire, is alluring.
Currently, five angiography FFR software solutions are available in Europe. Ninomiya et al compared these software systems in a prospective cohort in which all the analysis was conducted in a blinded, independent, academic core laboratory. For the study, the authors enrolled 355 patients with 444 target vessels. Of these, 390 met the criteria for optimal image quality required for analysis. Researchers performed FFR and iFR using an invasive pressure wire. The authors also analyzed quantitative coronary angiography (QCA) for each vessel.
Each angiographic FFR method bested two-dimensional QCA percentage diameter stenosis for the detection of clinically significant stenosis. However, the area under the curve (AUC) readings for detecting a positive invasive FFR (FFR ≤ 0.80) were in the range of 0.73 to 0.75, far lower than what had been reported in vendor-initiated trials. For iFR, the range was wider (between 0.68 and 0.74). False-positives were more common in the right coronary artery and the left circumflex, as well as in vessels with larger reference diameters. Conversely, false-negatives were more common in the left anterior descending and in small coronary vessels.
The authors concluded angio-derived FFR systems were useful and surpassed two-dimensional QCA techniques, but that its diagnostic accuracy did not match what has been reported by vendors.
COMMENTARY
Labeling of coronary plaque with a percent diameter stenosis is a time-honored practice in clinical cardiology. However, using this measurement as a surrogate for defining obstructive coronary disease is fraught with problems. Visual estimates of stenosis severity do not consider such important variables as plaque eccentricity, vessel ectasia, or length of the diseased segment as it relates to resistance to flow. On top of this, coronary angiography is an inherently subjective exercise. Show the same angiogram to 10 cardiologists, and one can expect multiple readings that are different in at least subtle ways.
QCA takes some of the subjectivity out of the equation, but it is subject to the same issues of stenosis vs. flow as routine cardiologist-read angiograms. In addition, QCA is time- and labor-intensive, and is used primarily in research. Invasive pressure wire-derived measurements are advantageous, but require the time, expense, and risks of placing a wire in coronary vessels for what often is a purely diagnostic exercise.
Using software algorithms to estimate FFR values from a standard angiogram shows great promise, as evidenced by the fact that at least five vendors have brought products to market. Still, wider clinical adoption has lagged. The Ninomiya et al study, performed in a blinded, independent lab, confirmed that angio-based FFR is feasible in most cases — but not in all. The authors noted problems such as vessel overlap, poor angiographic quality, table movement during cine angiographic acquisition, foreshortening, and heavy calcification. In addition, ostial lesions and vascular abnormalities (e.g., myocardial bridging) preclude optimal contrast opacification and present challenges for these techniques specifically.
Despite these difficulties, and the disappointingly low AUC (at least compared to the vendors’ reported figures) of angiographic FFR, we can expect these techniques to gain traction in the coming years, but not to replace invasive pressure-wire techniques completely. The advantages over routine two-dimensional angiography alone are evident here, even in the current iterations of these products.
With expected advances in the application of artificial intelligence techniques, both accuracy and processing time should continue to improve, leading to broader clinical adoption and availability. General cardiologists should be aware of these emerging techniques, as they will apply to patient care in coming years.
Using a blinded, independent core laboratory, researchers compared five commercially available angiographic fractional flow reserve (FFR) methods to the pressure wire-based FFR technique. The authors found the diagnostic accuracy of these software techniques surpassed that of two-dimensional quantitative coronary angiography, but fell short of what has been reported in validation trials run by individual vendors.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content