FFR Versus iFR in Assessment of Hemodynamic Lesion Significance

BACKGROUND AND RATIONALE FOR THE STUDY Coronary artery disease (CAD) is the most frequent causes of death and disability in developed countries. The main diagnostic method for detection of CAD is a coronary angiography (CAG. However, the correlation between CAG and proven hemodynamic significance in borderline stenosis (40-70% of lumen narrowing) is only about 50%. The main method for detection of flow-limiting lesions is fractional flow reserve (FFR). FFR is calculated as distal pressure (Pd) divided by proximal pressure (Pa). The cut-off for hemodynamic significance is 0.8 and less. The use of FFR for lesion assessment has a level of recommendation of I A in European guidelines for coronary interventions from 2014. The FAME trail revealed that routine usage of FFR measurement decreased the costs of coronary interventions over two years. The reasons are smaller quantity of implanted stents and decreased number of procedures for in stent restenosis as a direct result of the lower incidence of initial stent implantation. FAME-2 trial has shown lower incidence of urgent revascularization in patients with FFR less than 0,8 treated by coronary intervention compared to those treated conservatively. This study was preliminary halted before reaching mortality endpoint for safety reasons.

Critical condition for proper FFR measurement is maximal vasodilatation. Since we do not have marker for this condition based on pressure measurement only we can only presume that it was reached by adenosine administration. There are a lot of limitations for this assumption. Foremost, technique for proper intravenous and intracoronary adenosine administration must follow strict rules and therefore can be done incorrectly and leads to wrong result in same cases. Furthermore, impaired endothelial function leads to lower response to adenosine administration. This situation can leads to false negative FFR result. The patients with CAD have frequently significant endothelial dysfunction causing lower response to vasodilatation stimuli. Endothelial dysfunction will be analyzed by a system named EndoPAT (Itamar Medical, Israel) that measures ischemia induced vasodilatation on fingers. The hypothesis that has never been tested is how polymorphism in genes for enzymes playing an important role in proper endothelial function (HO-1, hemoxygenase-1 and ENOS, endothelial NO synthase) can influences adenosine-induced vasodilatation and subsequently FFR measurement. These polymorphisms are not infrequent in population with CAD (they can be found in 40-50% in such patients).

Another possible limitation for FFR measurement is type of blood flow in coronary arteries. In presence of coronary stenosis, plaque roughness or sharp angle of a lumen the type of flow can be changed from laminar to turbulent. This type of flow is even accelerated during hyperemia induced by adenosine administration. It leads to lose of energy and exaggerates pressure drop behind stenosis that - for this reason- may not be proportional to stenosis severity. This situation can theoretically causes false positive FFR measurement. To answer this question the investigators will analyze the endothelial shear stress (ESS) that can distinguish laminar and turbulent flow and morphological indices (plaque surface, plaque eccentricity, lumen volume and lumen shape) by optical coherence tomography (OCT).

The new index able to identify physiologically significant stenosis has been recently described. It is instantaneous wave-free ratio (iFR). This technique uses a pressure gradient as well, but -unlike to FFR- iFR compares pressures (proximal and distal to stenosis) only in a specific phase of diastole (so called "wave free period"), where microvascular resistance is naturally low and stable .

The cut off point for iFR is 0.9. The correlation between FFR and iFR is between 80% and 90% in all lesions. However, the correlation between FFR and iFR close to their cut-off point is only 50%-60%. It is substantial caveat, because to estimate hemodynamic significance of borderline lesions is the main indication for these techniques. Discrepancies between FFR and iFR can be very confusing a discouraging for lesion hemodynamic assessment.

Nowadays, FFR is the only invasive method for stenosis hemodynamic assessment supported in the guidelines. However, the iFR is faster, easier and cheaper, and for these reasons it can be used for more lesions in more patients, which can improve the situation of poor penetration of hemodynamic guidance for coronary intervention in daily practice. Moreover, an increasing number of evidences implies that iFR concept can be closer to real situation in coronary arteries (avoiding non-physiologic vasodilatation). It has been shown that FFR itself has a factor of variability of 10%. This means that around the cut-off points the FFR can be just as wrong as the iFR, perhaps even more so. Moreover, the latest study has shown that iFR provides better pressure-derived diagnostic agreement with CFR (coronary flow reserve) than FFR. The investigators have been developing, in close cooperation with The University of Iowa, new technique that improves the quality of flow measurement. It was successfully tested it in a animal trial. It is a novel way of intravascular Doppler signal processing resulting in more reliable velocity curve envelope acquisition. This software improvement allows instantaneous monitoring of real time microvascular resistance during any phase of cardiac cycle. Detection of during instantaneous wave free period can serve as verification of the proprietary iFR calculation. The system may also improve iFR measurement itself, because it can measure precisely in-phase with the lowest microvascular resistance. The measurement of resistance during wave free period is named iMR (instantaneous microvascular resistance) and it has never been measured during previous trials. This index can greatly help to distinguish which method (FFR or iFR) measures the pressure gradient during lower resistance.

Further target is to develop next generation of the software that can detect lowest microvascular resistance based on pressure measurement only, without necessity of using Doppler. It could offer very precise and simple methods for hemodynamic assessment of coronary lesions. This software must be tested and verified in a human study, which will be done in the General Teaching Hospital (where both intracoronary pressure and flow will be measured). Its clinical availability will be tested and verified by the international cooperating centers (where only pressure indices, FFR and iFR will be measured). The foreign centers will also examine coronary OCT that will be sent to corelab in The University of Iowa, collect blood samples for genetic analysis that will be done in Prague and, depend on their possibilities, also endothelial dysfunction by EndoPAT.

The University of Iowa, Iowa, USA will analyze OCT measurement and perform 3D vessel reconstruction. This university has a world known team for the unique 3D reconstruction of coronary arteries based on angiography and optical coherence tomography. This institution will do an automated analysis of plaque surfaces as well.

HYPOTHESES:

1. The level of microvascular resistance can be used to distinguish which type of measurement (FFR or iFR) was done during the lower and more stable stage of microvascular resistance. This comparison can possibly explain discrepancies between FFR and iFR measurements.
2. Based on new software, using pressure measurements only, it will be possible to automatically detect the time period with the lowest microvascular resistance. This could improve accuracy of both FFR and iFR measurements.
3. Plaque ruptures, erosions, irregularities of plaque geometry and plaque located near to bifurcations cause turbulence in blood flow. This accelerated pressure drop can lead to a false positive FFR.
4. Inadequate vasodilation caused by endothelial dysfunction can lead to false negative FFR.
5. Endothelial dysfunction can be more frequently found in patients with the risk type of polymorphism in genes playing an important role in vessel vasodilatation (ENOS, HO-1)

AIMS AND EXPECTED IMPACT ON CLINICAL PRACTICE

1. To use new software (developed at the author's workplace in cooperation with The University of Iowa) to determine which of two methods for the functional assessment of coronary stenosis (FFR and iFR) perform their measurements during a lower level of microvascular resistance. This software can measure microvascular resistance in real time.
2. To develop a new version of software for the detection of microvascular resistance level, based only on intracoronary pressures without flow analysis. This could substantially improve the accuracy of both pressure-based measurements, and potentially increase the correlation between pressure based measurements and flow-based CFR.
3. To study the potential influence of endothelial dysfunction and plaque morphology on discrepancies between FFR and iFR during functional assessments of coronary stenosis.
4. To study the influence of gene polymorphisms on endothelial dysfunction

METHODS Study design Patients with stable angina pectoris with suitable for coronary angiography will be suitable for the study. We plan to include 250 patients to the study (50 from General University Hospital in Prague and 200 from cooperating centers).

Functional examinations of coronary arteries. Coronary angiography will be performed as a first procedure for detection of severity and extent of coronary atherosclerosis. Stenosis between 40-80% (based on CAG) will be suitable for morphological and functional examinations. Combo wire (Volcano Corp., USA) for pressure and flow measurement will be introduced behind the lesion and basal indices will be measured during basal flow condition: basal flow speed, pressure gradient (Pd/Pa), iFR, Pd/Pa during lowest microvascular resistance proven by Doppler analysis, iMR. Adenosine will be administered either intracoronary as a bolus (240 ug) or in continual infusion (140 μg/kg/min) based on local practice. The hyperemic indices will be measured: maximal flow speed, CFR, FFR, HSR (hyperemic stenosis resistance. Cooperating centers will perform pressure measurements (FFR, iFR, Pd/Pa) only and they will send a raw data for further analysis (off-line calculation of iFR using the new software) in General Teaching Hospital in Prague.

Morphological examinations of coronary arteries Morphological assessment of the lesions will be done by OCT (St. Jude Medical, Inc). Catheter will be placed behind stenosis. Pullback will be done during flushing of contrast dye. OCT measurement will help to choose optimal treatment strategy by proper measuring of lumen size in a lesion and in the reference segments. PCI will be done according to local practice. OCT can be use after procedure for checking results, but this second examination is not part of the study.

More detail analyze will be done from 3D vessel reconstruction in The University of Iowa that is well know center for 3D coronary reconstruction. Scientific team from this university have received two US patents for 3D coronary reconstruction.

The methodology of 3 D reconstruction of coronary arteries The two-plane angiograms will be taken immediately prior to the pullback start and cover at least one heart cycle each. They will be used to extract the catheter path automatically along the expected pullback trajectory by a dynamic programming approach. From the known imaging geometry, an accurate 3-D model of the catheter path within the respective vessel segment is generated for end-diastolic heart phase. For OCT acquisition, motorized pullback ensures a constant pullback speed, thus allowing to asses each OCT image frame a specific location on the 3-D catheter trajectory model. The relative and absolute orientations of the OCT frames will be determined using previously reported system for establishing the absolute orientation in 3-D on IVUS (intravascular ultrasound) images. Visualization will be based on automated encoding of the derived contour data in VRML (Virtual Reality Modeling Language). Quantitative data can be derived from the contour data, such as luminal dimensions and plaque-cap thickness, actually considering the vessel curvature in contrast to conventional OCT reconstruction systems. The space between adjacent contours is interpolated to form a volume element. In locations of the plaque cap, integrating over an entire vessel segment or any part thereof yields the total plaque cap volume enclosed by the inner and outer cap surfaces. The quantification values can be included into the VRML model by color per vertex encoding, thus allowing an easy and fast visual assessment of the lesion or the results of the intervention by the physician.

Endothelial shear stress analysis. A steady flow computational fluid dynamics (CFD) analysis will be performed in the reconstructed arterial segments in order to analyze for the local fluid dynamic characteristics along the vessel segment.

Examination of endothelial dysfunction. Endothelial dysfunction will be measured by a system named EndoPAT (Itamar Medical, Israel). EndoPAT uses peripheral artery tone signal (PAT) for non-invasively measuring arterial tone changes in peripheral arterial beds17.

Genetic analysis of polymorphisms in gene for HO-1 and ENOS Patient's DNA will be isolated from peripheral blood leukocytes using standard techniques.

Statistical analysis Data will be prospectively stored in a database and will be processed using the software JMP®10.0.0, Copyright © 2012 SAS (Statistical analysis software) Institute Inc. (http://www.jmp.com) in collaboration with a professional statistician.