Author + information
- Published online May 1, 2017.
- aDepartments of Radiology and Medicine, Weill Cornell Medical College, New York, New York
- bDalio Institute of Cardiovascular Imaging, New York–Presbyterian Hospital, New York, New York
- ↵∗Address for correspondence:
Dr. James K. Min, New York–Presbyterian Hospital and Weill Cornell Medical College, Department of Radiology, 413 East 69th Street, Suite 108, New York, New York 10021.
In 1843, in his journals, the Danish theologian Søren Kierkegaard reflected: “It is perfectly true, as the philosophers say, that life must be understood backwards. But they forget the other proposition, that it must be lived forwards” (1). Applied to medicine, this statement asserts an importance to both the hindsight assessment of outcomes as well as the exploration of new experiences.
In cardiovascular imaging, new experiences often arise from the introduction of new technologies, with evidence development an enthusiastic focus of research. Fractional flow reserve derived from computed tomography (FFRCT) is one such technology. It has recently become available for U.S. Food and Drug Administration–approved clinical use as an alternative approach to traditional stress testing, with or without imaging, for the evaluation of patients with suspected ischemic heart disease. On the basis of a series of diagnostic performance studies—including DISCOVER-FLOW (Diagnosis of Ischemia-Causing Stenoses Obtained Via Noninvasive Fractional Flow Reserve), DeFACTO (Determination of Fractional Flow Reserve by Anatomic Computed Tomographic Angiography), and NXT (Analysis of Coronary Blood Flow Using CT Angiography: Next Steps)—FFRCT has been shown to have high accuracy compared with an invasive fractional flow reserve (FFR) reference standard, a measure of the differences in hyperemic pressure across a coronary lesion (2–4). The latter is an important metric, given its existence as the sole method for guiding decisions of coronary revascularization in a manner that improves event-free survival (5). Historical comparisons of FFRCT versus stress testing with imaging assert a superiority of FFRCT over these methods for the precise identification of coronary artery lesions that are ischemia-producing.
Beyond its diagnostic performance, FFRCT has also been evaluated for its ability to influence decisions of coronary revascularization over traditional metrics of coronary stenosis severity, as well as its ability to guide clinical decision-making compared with traditional approaches. For example, in the FFRCT RIPCORD (Does Routine Pressure Wire Assessment Influence Management Strategy at Coronary Angiography for Diagnosis of Chest Pain) trial, a multicenter substudy of the NXT trial, Curzen (6) evaluated the interpretation of coronary computed tomography angiography (CTA) studies from 3 interventional cardiologists, who reported their initial plan of action to prescribe optimal medical therapy, perform percutaneous coronary revascularization, or refer for coronary artery bypass surgery. These interventional cardiologists also specified which artery was to be treated. The investigators were then provided the results of the FFRCT, which, when combined with the coronary CTA results, were examined for their contribution to influence the proposed care strategy. The results indicated a significant rise in the prescription of optimal medical treatment alone, with a concomitant reduction in unnecessary invasive angiography combined with changes even in the vessels being targeted for revascularization. However, in large respect, the FFRCT RIPCORD study was hypothetical in nature, calling for a well-performed prospective study evaluating its concept.
More recently, the prospective multicenter international PLATFORM (Prospective Longitudinal Trial of FFRCT: Outcome and Resource Impacts) study was performed and findings presented at the European Society of Cardiology 2015 Late Breaking Clinical Trials session, and simultaneously published in the European Heart Journal (7). The PLATFORM investigators evaluated nearly 600 patients at 11 sites who presented with symptomatic coronary artery disease (CAD) and who were prospectively evaluated by using standard of care approaches versus those involving the combination of coronary CTA and FFRCT. In this study, a 61% reduction in normal invasive angiograms was observed with the use of the coronary CTA/FFRCT diagnostic paradigm, suggesting a higher specificity (i.e., a lower rate of false-positive study results). The “real-world” cancellation of invasive testing did not adversely affect the low rate of clinical events observed in both the standard of care and the coronary CTA/FFRCT groups. A subsequent publication of the PLATFORM study by Hlatky et al. (8) reported a lower cost of care for the patients being managed with coronary CTA/FFRCT over the standard approach, by more than $3,000 per patient for patients being referred for invasive coronary angiography. Associated with this cost reduction was a superior improvement in quality of life—both overall and angina-specific—with the coronary CTA/FFRCT approach in patients being referred for noninvasive imaging.
Based on the high reported diagnostic performance of coronary CTA/FFRCT, the apparent efficacy of coronary CTA/FFRCT to influence clinical decision-making, the reduction in unnecessary referral to angiography, lower costs, and the higher quality of life, Nørgaard et al. (9) became early adopters of this technology. In this issue of iJACC, they report the clinical outcomes of a true real-world assessment of coronary CTA/FFRCT during its first 12-month period of clinical use. In this early stage of adoption, the investigators applied FFRCT to 189 patients who were referred for coronary CTA (15%). Despite the real-world nature of this study, with its expected nonnegligible rate of impaired diagnostic image quality often present in daily clinical practice, conclusive FFRCT results were achievable in 98% of patients. On the basis of these FFRCT findings, invasive angiography was performed in approximately one-third of patients, and FFR was performed in approximately 20% of these patients. Correct classification of FFRCT was observed in approximately 3 of 4 patients undergoing invasive evaluation. Among the majority of patients being deferred from invasive angiographic assessment on the basis of the FFRCT findings, no adverse clinical events occurred during the 12-month follow-up.
These findings (9) are of high import with respect to both the evaluation and actual widespread clinical adoption of FFRCT in daily practice. As advocated by the philosopher Kierkegaard, by early adoption, Nørgaard et al. (9) have embraced CAD evaluation by FFRCT “forwards” but have reported their understanding of it “backwards,” and they offer several important points of evidence. First, this study reports the findings of a very early stage in use of FFRCT that may not necessarily reflect its future use, even among these investigators who are expert and familiar with the technology. This situation is reflected in the nearly 20% of patients who ostensibly underwent “confirmatory” invasive FFR. Importantly, the findings of FFRCT positivity in 3 of 4 patients confirmed by using invasive FFR compares favorably to stress testing, with reported obstructive anatomic CAD lesions in approximately 1 of 3 patients (10). Pertaining to the latter, an invasive FFR evaluation would likely reduce this number even further. Despite the favorable ratio, FFRCT may suffer from its own precision. Unlike stress perfusion imaging, which does a generally poor job of specifying ischemia-causing coronary lesions, FFRCT precisely identifies the location of hyperemic pressure drop. As such, any discordance with invasive FFR may be considered a false-positive finding, with the inability to test such a concept by using stress testing. It remains likely that increased clinical experience will inform the use of FFRCT use, and the “confirmatory” invasive FFR findings will fall. However, this prediction still requires proof.
Importantly, the patient population studied in this report (9) are of a generally lower risk than has been reported for previous myocardial perfusion studies, with a younger age, lower prevalence of male subjects, and lower rates of traditional CAD risk factors (e.g., diabetes, hypertension, hyperlipidemia). The predominance of symptoms consistent with atypical angina, coupled with a Diamond-Forrester pre-test probability of 34%, reinforce this notion. Whether the results of this study can be applied to a higher risk population or to patients with known previous CAD, which is a population not yet indicated for clinical use by FFRCT, remains to be seen.
Furthermore, the quality of coronary CTA in this study by Nørgaard et al. (9) requires consideration. The accuracy of FFRCT is invariably influenced by the quality of the coronary CTA study. In this report, the authors report a nearly 100% acceptance rate of coronary CTA for FFRCT calculation. This information is valuable because it establishes the feasibility of performing coronary CTA to a quality that uniformly is acceptable for FFRCT calculation. However, the high quality of coronary CTA offered by these expert investigators likely does not represent the current quality attainable at most sites. One potential solution may be the interaction of site clinicians performing coronary CTA and the vendors providing FFRCT services. This approach might offer an opportunity to organize a coronary CTA-FFRCT team that can perform real-time quality control and quality improvement initiatives aimed at not only the acceptance rate of coronary CTAs for FFRCT calculation but overall improvement of coronary CTA imaging.
Despite the generally large numbers of patients evaluated by Nørgaard et al. (9), numerous limitations must be emphasized, many of which have been acknowledged by the investigators. Of perhaps the greatest importance is the single-center observational nature of the study, a limitation of any trial at a sole site but particularly relevant in this study’s findings. Expertise, early adopter status, and general site acceptance of FFRCT utility invariably affects the current study findings, and care should be made not to indiscriminately apply them to all sites considering use of FFRCT. Furthermore, similar to what would be expected at other sites, the use of FFRCT was guided by the clinicians caring for the tested patients. Although similar to the use of stress testing and reflective of the “real-world” design, physician discretion in this study was guided neither by a standardized approach nor by societal practice guidelines. The latter of these would be highly valuable to guide the widespread use of FFRCT but are currently unavailable. The current report should help, in some regard, to frame the discussion of such evidence-based documents.
Looking backward and acting forward, Nørgaard et al. (9) should be recognized for providing the field with important information regarding the use of FFRCT in the “real world.” Although certain limitations exist, which somewhat dampen the potential generalizability of the study findings to the “rest of the world,” they nevertheless offer essential information that poise the combination of coronary CTA and FFRCT to transcend current practice paradigms into a more diagnostically specific and accurate method of CAD evaluation. The goal of this evaluation is to more precisely identify those individuals who may benefit from referral for invasive assessment and therapy versus those who may not.
↵∗ Editorials published in JACC: Cardiovascular Imaging reflect the views of the authors and do not necessarily represent the views of JACC: Cardiovascular Imaging or the American College of Cardiology.
Dr. Min has served as a consultant to GE Healthcare, HeartFlow, and the Cardiovascular Research Foundation; is on the advisory board for Arineta; has received grants from the National Institutes of Health (R01 HL111141, R01 HL 115150, and R01 HL 118019); has received a gift from the Dalio Foundation; and has ownership in MDDX and Autoplaque.
- 2017 American College of Cardiology Foundation
- ↵Kierkegaard S. Journal IV A 164 (1843). Available at: http://quotes.yourdictionary.com/author/quote/606433. Accessed April 1, 2016.
- Koo B.K.,
- Erglis A.,
- Doh J.H.,
- et al.
- Nørgaard B.L.,
- Leipsic J.,
- Gaur S.,
- et al.,
- NXT Trial Study Group
- ↵Curzen N. The FFRCT RIPCORD study: does the routine availability of computed tomography derived FFR influence management strategy of patients with stable chest pain compared to computed tomographic angiography alone? Presented at: EuroPCR; May 19, 2015; Paris, France.
- Douglas P.S.,
- Pontone G.,
- Hlatky M.A.,
- et al.,
- PLATFORM Investigators
- Hlatky M.A.,
- De Bruyne B.,
- Pontone G.,
- et al.,
- PLATFORM Investigators
- Nørgaard B.L.,
- Hjort J.,
- Gaur S.,
- et al.