Does Diabetes Affect Angiographically Derived (QFR) Translesional Physiology?: Looking at the FAVOR III Diabetic Subset
Fractional flow reserve (FFR) has become the criterion standard for the invasive assessment of coronary ischemia, based on multiple validations against noninvasive tests and clinical outcome trials. FFR is calculated from the ratio of translesional distal coronary pressure (Pd) to aortic pressure (Pa) as directly measured from a coronary pressure wire during adenosine-induced hyperemia. Because of the side-effects, cost, and time required of using adenosine, a group of nonhyperemic resting pressure ratios (NHPRs) were subsequently developed, found to be noninferior to FFR, and rapidly adopted in clinical practice. Following a wave of evolving studies of coronary physiology, the use of 3-dimensional (3D) angiography with sophisticated computational fluid dynamics can now generate angiographically derived FFR.
The use of angiographically derived FFR will likely meet or exceed pressure wire–based FFR in catheterization laboratories across the globe within the next few years. Just as the NHPRs (instantaneous wave–free ratio, diastolic pressure ratio, resting full cycle ratio, and diastolic hyperemia free ratio, as well as resting translesional pressure ratio) have quickly risen to daily use alongside FFR, the angiographically derived FFR methods, such as quantitative flow ratio (QFR) (Medis), virtual FFR (Pie Medical Imaging), and FFRangio (CathWorks), have overcome early concerns about ease of use, validation, and fundamental limitations from 2-dimensional angiography. Outcome studies for QFR-guided percutaneous coronary intervention (PCI) have recently demonstrated superiority over angiographically guided PCI in the same way as have the FAME (Fractional Flow Reserve Versus Angiography for Multivessel Evaluation) studies and other large physiologically guided PCI trials.
In this issue of the Journal of the American College of Cardiology, Jin et al address the impact of diabetes mellitus (DM) on QFR-guided PCI by examining the diabetic subgroup from the FAVOR III China trial. The main result of FAVOR III was that in 3,825 randomized patients, QFR-guided PCI was superior to angiographically guided PCI in terms of 1-year major adverse cardiovascular events (MACE) (a composite of death, myocardial infarction, or ischemia-driven revascularization). The Jin et al substudy encompassed the 1,295 subjects (34%) who had DM, 347 (27%) of whom were insulin dependent. Compared with the angiography-based approach, the QFR-guided strategy consistently and significantly reduced 1-year MACE in both diabetic patients (6.2% vs 9.6%); and nondiabetic patients (5.6% vs 8.3%). In patients in whom PCI was deferred based on QFR, there was no difference in the 1-year MACE rate in patients with and without DM (4.5% vs 6.2%; P = 0.51). In short, the study demonstrates that a QFR-guided lesion selection strategy improves PCI outcomes compared with angiography guidance regardless of DM status.
Such a result was not a foregone conclusion, because DM is known to impair microcirculatory function, which potentially interferes with the hyperemia induction needed for accurate FFR measurements, and theoretically with the contrast frame count used to estimate contrast flow velocity for the QFR calculation. Also, DM is more often associated with diffuse disease and small-vessel disease, which may affect the QFR calculation.
It is also well known that DM alters the natural history of coronary artery disease. Insulin-dependent DM is a notoriously bad actor. The mechanisms by which DM increases adverse event rates are multifactorial but related to glucose dysregulation, metabolic syndromes, inflammation, complex atherosclerosis with diffuse epicardial as well as small (<400 μm) vessel involvement with variable degrees of microvascular dysfunction. These conditions as well as comorbidities of renal dysfunction and other microvascular angiopathy confound predictions on outcomes. In the DEFINE-FLAIR (Functional Lesion Assessment of Intermediate Stenosis to Guide Revascularisation) and iFR-SWEDEHEART (Evaluation of iFR vs FFR in Stable Angina or Acute Coronary Syndrome) studies, patients with DM had a significantly higher risk of MACE than those without DM, whether they received iFR-guided or FFR-guided treatment.
Furthermore, DM complicates the prognostic capabilities of FFR (and likely QFR) owing to factors other than a nonischemic value (eg, FFR >0.80). Kennedy et al found that patients with IDDM and previous revascularization had a higher incidence (18.1% vs 7.5%; P < 0.01) of deferred lesion failure compared with non-DM patients. Barbato et al also noted that patients at low risk according to coronary hemodynamics (mean FFR = 0.88), but at high risk due to clinical factors (insulin-dependent DM, previous revascularization, previous myocardial infarction) had lesion failure in 12% of lesions, with 9% of all deferred lesions resulting in acute coronary syndrome.
There are a few limitations to the present study and to QFR in general that are worth mentioning. Jin et al concede that as in all subgroup studies the interpretation of the significance of low-frequency end points should come from a sufficiently powered study group. Although the subgroup was prespecified, caution is needed. Furthermore, data on the detailed type of DM, insulin and medication use, and status of HbA1c were unavailable, likely resulting in some mixing of clinical severity of the comorbidities. DM is not a single entity, and its clinical impact produces variable outcomes independently from the measured translesional physiology, whether directly with a pressure wire or angiographically derived.
Intravascular imaging for PCI was infrequently used. How much this loss might have affected outcomes is unknown but would likely be similar in both groups if used routinely. Finally, the authors rightly acknowledge that these results should not be extrapolated to patients with complex coronary anatomy or acute coronary syndromes. Of course, longer-term follow-up for both groups will be helpful.
Regarding QFR accuracy, several considerations in computational assumptions should be noted. QFR software assumes that coronary pressure remains constant through normal epicardial arteries, and a simple quadratic equation using coefficients derived from flow data is used to calculate the pressure drop across a stenosis, based on the geometry characterized by 3D quantitative coronary angiography (QCA). The computation is therefore only as good as the angiograms used for input. QFR further assumes that coronary flow velocity distal to the stenosis is similar to that proximal to the stenosis, and the mass flow rate along the vessel can be calculated using the mean flow velocity (Thrombolysis In Myocardial Infarction frame count) and 3D-QCA anatomic parameters. With these assumptions, QFR studies have shown good correlation and diagnostic accuracy with wire-based FFR.
The Bottom Line and the Future
The FAVOR III study was the largest randomized coronary physiology outcome study performed to date, larger than the pivotal FAME, FAME2, DEFINE-FLAIR, and iFR-SWEDEHEART trials. The study size provides adequate power for subgroup analysis, and the present statistically significant results are reassuring that the overall FAVOR III conclusions remain valid in diabetic patients. In short, QFR, like FFR, is not affected by DM. QFR-guided intervention, like FFR-guided intervention, is superior to angiographically guided intervention in terms of MACE.
Advances in coronary physiology have occurred at an increasing and breathtaking speed, as demonstrated by the number and rate of publications on the subject (Figure 1). Many of us have watched FFR struggle for acceptance for 20 years, but clinical uptake did not occur until well after the publication of the FAME and FAME2 studies in 2012. Each advance in the science has appropriately built on the previous work. The creators of iFR leveraged the FFR outcome data to demonstrate that noninferiority to FFR would be clinically useful. In contrast to the decades for FFR, the NHPRs have taken only 5 years to reach significant clinical adoption, owing to the prior acceptance of coronary physiology along with the simplicity of not needing to induce hyperemia. Similarly, QFR in a short time has already successfully achieved all the milestones (theoretical/basic science, statistical correlation with FFR, and now large outcome studies) that iFR did. The advantages of angiographic FFR (comprehensive 3-vessel analysis, time savings, and avoidance of the pressure wire and hyperemia) will undoubtedly drive adoption. In the very near future, operators will use angiographically derived FFR on a routine basis and keep the pressure wire as the fail-safe for situations that require the highest level of confidence in their interventional decision making.
Yearly Number of Publications for FFR, iFR, and QFR
Graphic of the number of publications for FFR, iFR (Philips.), and QFR (Medis) by year from 1995 to 2020. The rate of papers published for FFR remained low until the publication of the FAME 2 trial in 2012. Data from PubMed, 2022. FFR = fractional flow reserve; IFR = instantaneous wave–free ratio; QFR = quantitative flow ratio.
This article is reproduced from JACC journals.
Address: Room 201, 2121 Hongmei South Road, Minhang District, Shanghai
Address: room 709, No.8, Qihang international phase III, No.16, Chenguang East Road, Fangshan District, Beijing
Tel: 13331082638（Liu Jie）
Address: No. 15, Longrui street, longguicheng, Taihe Town, Baiyun District, Guangzhou