Statistics from Altmetric.com
Coronary angiography remains far and away the most accurate morphologic assessment of the lumen of the epicardial coronary arteries. Although non-invasive imaging is rapidly advancing, the temporal and spatial resolution of coronary angiography is still unsurpassed and it will therefore remain as a roadmap for interventional cardiologists and cardiac surgeons. Most non-invasive cardiologists still consider a coronary angiogram as “invasive” (although this is a debatable classification in 2008), but welcome the high level of anatomical information. However, in contrast to its topographical precision, angiography is limited in gauging the functional repercussions of coronary stenoses. Yet, functional severity of atherosclerotic narrowings is the single most important prognostic factor in patients with documented coronary artery disease. This was recently highlighted in several large meta-analyses: the estimated annualised rate of myocardial infarction or cardiac death was approximately 0.5% per year after a normal myocardial perfusion imaging or a normal stress echocardiogram in patients with unknown coronary anatomy.1 w1 In addition, the DEFER trial showed that the annual rate of death or myocardial infarction is approximately 1% in patients with angiographically documented epicardial coronary stenoses that are not functionally significant.2 In daily practice, however, non-invasive testing is performed in a minority of patients undergoing angioplasty.w2 Since the angiographic degree of stenosis is a poor tool to establish the functional significance of a given stenosis, a large number of inappropriate decisions are taken regarding revascularisation.3 Thus, the combination of highly accurate anatomic assessment and precise functional information is indispensable to tailor the treatment of patients with suspected or known coronary artery disease. Accordingly, the combination of coronary angiography and pressure derived fractional flow reserve in the catheterisation laboratory emerges as the only true “all-in-one” approach as it combines anatomy, physiology, and the possibility of “ad hoc” treatment.
The present review will focus on coronary pressure derived fractional flow reserve.
Box 1 Simplified theoretical explanation illustrating how a ratio of two flows can be derived from a ratio of two pressures provided these pressures are recorded during maximal hyperaemia
1. Fractional flow reserve (FFR) is the ratio of hyperaemic myocardial flow in the stenotic territory (Qsmax) to normal hyperaemic myocardial flow(QNmax)
2. Since the flow (Q) is the ratio of the pressure (P) difference across the coronary system divided by its resistance (R), Q can be substituted as following:
3. Since the measurements are obtained under maximal hyperaemia, resistances are minimal and therefore equal, and thus they cancel out:
4. In addition Pv is negligible as compared to Pa or Pd, therefore:
Pa, aortic pressure; Pd, distal coronary pressure; Pv, venous pressure; QSmax, hyperaemic myocardial blood flow in the stenotic territory; QNmax, hyperaemic myocardial blood flow in the normal territory; RSmax, hyperaemic myocardial resistance in the stenotic territory; RNmax, hyperaemic myocardial resistance in the normal territory.
DEFINITION OF FRACTIONAL FLOW RESERVE (FFR)
Fractional flow reserve (FFR) is the ratio of maximal myocardial blood flow in the case of a diseased artery to maximal myocardial blood flow if that same artery were to be normal.4 It may be clear that FFR is a ratio of two flows: the maximal myocardial flow in the stenotic territory divided by the maximal myocardial flow in the same territory in the hypothetical case that the stenosis was absent.5 Stated another way, FFR represents the extent to which maximal myocardial blood flow is limited by the presence of an epicardial stenosis. If FFR is 0.60, it means that maximal myocardial blood flow reaches only 60% of its normal value. Conversely, FFR provides the interventionalist with the exact extent to which optimal stenting of the epicardial stenosis will increase maximal myocardial blood flow. An FFR of 0.60 implies that stenting the focal stenosis responsible for this abnormal FFR should bring FFR to 1.0, which represents an increase in maximal myocardial blood flow of 67%.
FROM PRESSURE TO FLOW
FFR is a ratio of two flows. It has been shown, however, that this ratio of two flows can be derived from two pressures provided they are both measured during maximal hyperaemia. The theoretical explanation of this relationship between hyperaemic flows and hyperaemic pressures is displayed in box 1 and fig 1.
PRACTICAL ASPECTS OF FFR MEASUREMENTS
The use of diagnostic catheters is technically feasible.w3 Yet, due to higher levels of friction hampering wire manipulation, the smaller internal calibre prejudicing pressure measurements, and the inability to perform ad hoc percutaneous coronary intervention (PCI) using diagnostic catheters, the use of guiding catheters is recommended.
Measuring intracoronary pressure requires the use of a specific solid state sensor mounted on a floppy-tipped guide wire. In mainstream practice two such systems exist, namely the PressureWire (RadiMedical Systems Inc, Uppsala, Sweden) and the Volcano WaveWire (Volcano Inc, Rancho Cordova, California, USA). The sensor is located at the junction between the 3 cm long radiopaque tip of the wire and the remainder of the wire. The last generations of these 0.014 inch wires have similar handling characteristics to most standard angioplasty guide wires. Before introducing the sensor into the vessel to be studied, the pressures recorded by the sensor and by the guiding catheter should be equalised.
As soon as any device is advanced into the coronary tree, the use of the same anticoagulation regimens as employed during a PCI procedure is recommended: heparin adjusted to weight, validated by a monitored activated coagulation time (ACT) of at least 250 s.
It is absolutely essential to induce maximal vasodilatation of the two compartments of the coronary circulation (epicardial or “conductance arteries” and the microvasculature or “resistance arteries”). The pharmacological options for inducing hyperaemia are summarised in table 1.w4–6
A bolus of 200 μg isosorbide dinitrate (or any other form of intracoronary nitrates) allows the abolition of any form of epicardial vasoconstriction.
Microvascular vasodilation is equally paramount for the calculation of FFR. Gauging pressure differences at rest does not offer a definitive measure—we cannot emphasise too strongly that there is no such thing as a “baseline FFR”. Even when the resting pressure gradient is large we recommend the induction of hyperaemia because it allows us to evaluate what is the residual resistance reserve. An example of a typical coronary pressure tracing during the administration of intravenous adenosine is shown in fig 2.
SPECIAL FEATURES OF FFR
FFR has a number of unique characteristics that make this index particularly suitable for functional assessment of coronary stenoses and clinical decision making in the catheterisation laboratory.
FFR has a theoretical normal value of 1 for every patient, for every artery and for every myocardial bed
An unequivocally normal value is easy to refer to but is rare in clinical medicine. Since in a normal epicardial artery there is virtually no decline in pressure, not even during maximal hyperaemia,6 it is obvious that Pd/Pa will equal or be very close to unity. This means that normal epicardial arteries do not contribute to the total resistance to coronary blood flow. The lowest value found in a total of 65 strictly normal coronary arteries was 0.92.6 7 Yet it is important to realise that in normal looking coronary arteries in patients with proven atherosclerosis elsewhere, the epicardial coronary arteries may contribute to total resistance to coronary blood flow even though there is no discrete stenosis visible on the angiogram. In approximately 50% of these arteries, FFR is lower than the lowest value found in strictly normal individuals. In approximately 10% of atherosclerotic arteries, FFR will be lower than the ischaemic threshold.6 Practically speaking, this finding implies that myocardial ischemia might be present in atherosclerotic patients in the absence of discrete stenoses.
FFR has a well defined cut-off value with a narrow “grey zone” between 0.75 and 0.80
Cut-off or threshold values are values that distinguish normal from abnormal levels for a given measurement. To enable adequate clinical decision making in individual patients it is paramount that any level of uncertainty is reduced to a minimum. Stenoses with an FFR <0.75 are almost invariably able to induce myocardial ischaemia, while stenoses with an FFR >0.80 are almost never associated with exercise induced ischaemia (table 2). This means that the “grey zone” for FFR (between 0.75–0.80) spans over <10% of the entire range of FFR values.
FFR is not influenced by systemic haemodynamics
In the catheterisation laboratory systemic pressure, heart rate and left ventricular contractility are prone to change. In contrast to many other indices measured in the catheterisation laboratory, changes in systemic haemodynamics do not influence the value of FFR in a given coronary stenosis.8 In addition, FFR measurements are extremely reproducible.w7 This is due not only to the fact that aortic and distal coronary pressures are measured simultaneously, but also to the extraordinary capability of the microvasculature to repeatedly vasodilate to exactly the same extent. These characteristics contribute to the accuracy of the method and to the trust in its value for clinical decision making.
FFR takes into account the contribution of collaterals
Whether myocardial flow is provided antegradely by the epicardial artery, or retrogradely through collaterals, does not really matter for the myocardium. Distal coronary pressure during maximal hyperaemia reflects both antegrade and retrograde flow according to their respective contribution. This holds for the stenoses supplied by collaterals but also for stenosed arteries providing collaterals to another more critically diseased vessel. Figure 3 shows the influence on the FFR measurements of left to right collaterals.
FFR specifically relates the severity of the stenosis to the mass of tissue to be perfused
The larger the myocardial mass subtended by a vessel, the larger the hyperaemic flow, and in turn, the larger the gradient and the lower the FFR. This explains why a stenosis with a minimal cross sectional area of 4 mm2 has a totally different haemodynamic significance in the proximal left anterior descending artery (LAD) versus the second marginal branch.
FFR has unequalled spatial resolution
The exact position of the sensor in the coronary tree can be monitored under fluoroscopy, and documented angiographically. Pulling back the sensor under maximal hyperaemia provides the operator an instantaneous assessment of the abnormal resistance of the arterial segment located between the guide catheter and the sensor. While other functional tests reach a “per patient” accuracy (exercise ECG) or, at best, a “per vessel” accuracy (myocardial perfusion imaging), FFR reaches a “per segment” accuracy with a spatial resolution of a few millimetres.
Fractional flow reserve: key points
Fractional flow reserve (FFR) is the extent to which maximal myocardial flow is limited by the presence of an epicardial stenosis.
The normal value of FFR is 1 whatever the patient, the territory, the systemic haemodynamics, the size of the vessel or the myocardial mass.
The FFR values of 0.75 to 0.80 distinguish significant from non-significant stenoses.
The FFR value takes into account the contribution of collaterals.
APPLICATION OF FFR IN DIFFERENT LESION SUBSETS
FFR in angiographically dubious stenoses
Cardiologists have a large glossary of terms to describe coronary narrowings with uncertain functional consequences: mild-to-moderate stenoses, dubious lesions, intermediate stenoses, non-flow limiting, non-significant stenoses, etc. The multiplicity of these denominations betrays their inaccuracy. The main general indication for FFR is, and will remain, the precise assessment of the functional consequences of a given coronary stenosis with unclear haemodynamic significance.9 In a study of 45 patients with angiographically dubious stenoses it was shown that FFR has a much larger accuracy in distinguishing haemodynamically significant stenoses than exercise ECG, myocardial perfusion scintigraphy and stress echocardiography taken separately. Furthermore, the results of these non-invasive tests are often contradictory which renders appropriate clinical decision making difficult (fig 4). In addition, the clinical outcome of patients in whom PCI has been deferred, because the FFR indicated no haemodynamically significant stenosis, is very favourable. In this population the risk of death or myocardial infarction is approximately 1% per year, and this risk is not decreased by PCI.2 These results strongly support the use of FFR measurements as a guide for decision making about the need for revascularisation in “intermediate” lesions. Figure 5 illustrates how two angiographically similar stenoses may have a completely different haemodynamic severity. One of them should be revascularised, the other not. Based solely on the angiogram, the decision should be identical in both cases, which would lead to an inappropriate interventional decision in one of these patients.
FFR in left main stem disease
The presence of a significant stenosis in the left main stem is of critical prognostic importance.w8 Conversely, revascularisation of a non-significant stenosis in the left main may lead to atresia of the conduits, even when internal mammary arteries are used.w9 Furthermore, the left main is among the most difficult segments to assess by angiography.w10 Non-invasive testing is often non-contributive in patients with a left main stenosis. Perfusion defects are often seen in only one vascular territory, especially when the right coronary artery is significantly diseased.w11 In addition tracer uptake may be reduced in all vascular territories (“balanced ischaemia”) giving rise to false negative studies.10 Several studies have shown that FFR could be used safely in left main stenosis and that the decision not to operate on left main stenosis with an FFR >0.75 is safe.w12–14 In addition, angiographic assessments of left main lesions with an FFR <0.75 were no different from those with an FFR >0.75, further reinforcing the importance of physiological parameters in case of doubt. Therefore, patients with an intermediate left main stenosis deserve physiologic assessment before blindly taking a decision about the need for revascularisation. Two examples shown in fig 6 illustrate how FFR measurements in the left main may drastically influence the type of treatment in these patients.
Left main disease is rarely isolated. When tight stenoses are present in the LAD or in the left circumflex artery (LCx) the presence of these lesions will tend to increase the FFR measured across the left main. The influence of an LAD/LCx lesion on the FFR value of the left main will depend on the severity of this distal stenosis but, even more, on the vascular territory supplied by this distal stenosis. For example, if the distal stenosis is in the proximal LAD, its presence will notably impact the stenosis in the left main. If the distal stenosis is located in a small second marginal branch, its influence on the left main stenosis will be minimal (fig 7).
FFR in multivessel disease
Patients with “multivessel disease” actually represent a very heterogeneous population. Their anatomical features (number of lesions, their location, and their respective degree of complexity) may vary tremendously and have major implications for the revascularisation strategy. Moreover, there is often a large discrepancy between the anatomic description and the actual severity of each stenosis. For example, a patient may have “three vessel disease” based on the angiogram, but actually have only two haemodynamically significant stenoses; vice versa, a patient can be considered as having one vessel disease of the right coronary artery (RCA) but actually have a haemodynamically significant stenosis of the left main. Figure 8 shows a typical example of a patient in whom the RCA and the LCx are critically narrowed and in whom the mid-LAD shows a mild stenosis. Myocardial perfusion imaging showed a reversible perfusion defect in the inferolateral segments and a normal flow distribution in the segments supplied by the LAD. In contrast, FFR shows that all three vessels are significantly narrowed but to a different extent. This has a major implication as far as revascularisation is concerned. Preliminary FFR guided revascularisation strategies in patients with multivessel disease were very encouraging.11 w15 w16 Tailoring the revascularisation according to the functional significance of the stenoses rather than on their mere angiographic appearance may decrease costs and avoid the need for surgical revascularisation. A randomised prospective trial testing the hypothesis that FFR might usefully guide percutaneous revascularisation in patients with multivessel coronary artery disease is being performed.w17
FFR after myocardial infarction
After a myocardial infarction, previously viable tissue is partially replaced by scar tissue. Therefore, the total mass of functional myocardium supplied by a given stenosis in an infarct related artery will tend to decrease.12 By definition, hyperaemic flow and thus hyperaemic gradient will both decrease as well. Assuming that the morphology of the stenosis remains identical, FFR must therefore increase. This does not mean that FFR underestimates lesion severity after myocardial infarction. It simply illustrates the relationship that exists between flow, pressure gradient and myocardial mass and, conversely, illustrates that the mere morphology of a stenotic segment does not necessarily reflect its functional importance. This principle is illustrated in fig 9. Recent data confirm that the hyperaemic myocardial resistance in viable myocardium within the infarcted area remains normal.w18 This further supports the application of the established FFR cut-off value in the setting of partially infarcted territories. Earlier data had suggested that microvascular function was abnormal in regions remote from a recent myocardial infarction.w19 w20 However, more recent work taking into account distal coronary pressure indicates that hyperaemic resistance is normal in these remote segments.w21 These data support the use of FFR to evaluate stenoses remote from a recent myocardial infarction.
FFR in diffuse disease
Histopathology studies and, more recently, intravascular ultrasound have shown that atherosclerosis is diffuse in nature and that a discrete stenosis in an otherwise normal artery is actually rare. The concept of a focal lesion is a mainly angiographic description but does not reflect pathology. Until recently, it was believed that when no focal narrowing of >50% was seen at the angiogram no abnormal resistance was present in the epicardial artery. It was therefore assumed that distal pressure was normal and thus that “diffuse mild disease without focal stenosis” could not cause myocardial ischemia. This paradigm has recently been shifted: the presence of diffuse disease is often associated with a progressive decrease in coronary pressure6 and flow,13 and this cannot be predicted from the angiogram. In contrast this decline in pressure correlates with the total atherosclerotic burden.14 In approximately 10% of patients this abnormal epicardial resistance may be responsible for reversible myocardial ischaemia. In these patients chest pain is often considered non-coronary because no single focal stenosis is found, and the myocardial perfusion imaging is wrongly considered false positive (“false false positive”).w22 Such diffuse disease and its haemodynamic impact should always be kept in mind when performing functional measurements. In a large multicentre registry of 750 patients FFR was obtained after technically successful stenting. A post-PCI FFR value <0.9 was still present in almost one third of patients and was associated with a poor clinical outcome.15 The only way to demonstrate the haemodynamic impact of diffuse disease is to perform a careful pull-back manoeuvre of the pressure sensor under steady state maximal hyperaemia (fig 10).
FFR in sequential stenoses
When several stenoses are present in the same artery, the concept and the clinical value of FFR is still valid to assess the effect of all stenoses together. Yet, it is important to realise that when several discrete stenoses are present in the same coronary artery, each of them will influence hyperaemic flow and therefore the pressure gradient across the other one. The influence of the distal lesion on the proximal is more important than the reverse. The FFR can theoretically be calculated for each stenosis individually.16 w23 However, this is neither practical nor easy to perform and therefore of little use in the catheterisation laboratory. Practically, as for diffuse disease, a pull-back manoeuvre under maximal hyperaemia is the only way to appreciate the exact location and physiological significance of sequential stenoses.
FFR in bifurcation lesions
Overlapping of vessel segments as well as radiographic artefacts render bifurcation stenoses particularly difficult to evaluate at angiography, while PCI of bifurcations is often more challenging than for regular stenoses. The principle of FFR guided PCI applies in bifurcation lesions even though clinical outcome data are currently limited. Two recent studies by Koo et al.17 w24 used FFR in the setting of bifurcation stenting. The results of these studies can be summarised as follows. (1) After stenting the ostium of the side branch looks often “pinched”. Yet such stenoses are grossly overestimated by angiography—none of these ostial lesions where the diameter stenosis was estimated as <75% were found to have an FFR below 0.75. (2) When kissing balloon dilation was performed only in ostial stenoses with an FFR <0.75, the FFR at 6 months was >0.75 in 95% of all cases.
CAVEATS AND DIFFICULT SITUATIONS
FFR and left ventricular hypertrophy
Reversible myocardial ischaemia might be observed in patients with left ventricular hypertrophy even in the absence of epicardial stenosis. In such patients these (subendocardial) flow maldistributions are related to a mismatch between capillary density and the volume of the myofilaments. Therefore, the usual cut-off values should be used with caution. In patients with an FFR <0.75, ischaemia will uniformly be present; in patients with FFR values between 0.75–0.90 it is more hazardous to include reversible ischaemia.
Fractional flow reserve: key points
Maximal hyperaemia is paramount for the concept of FFR (there is no such thing as a “resting FFR”)
Slow pullback of the sensor under steady state hyperaemia provides information about coronary artery haemodynamics with unequalled spatial resolution.
Functional information is critical to determine which patients and which stenoses deserve revascularisation. FFR is able to provide this information in the catheterisation laboratory where PCI can be performed when appropriate.
FFR and acute myocardial infarctions
During primary PCI for acute myocardial infarction, the combination of the symptoms, the ECG and the angiogram make it possible to determine the culprit lesion in the vast majority of cases. In addition, thrombus embolisation, myocardial stunning, acute ischaemic microvascular dysfunction, among other factors, make reaching complete microvascular dilatation unlikely. Finally, these changes are dynamic in nature and measurements obtained during the acute phase might well be strikingly different from those obtained 1 day later. Therefore, measuring FFR in the culprit lesion makes very little sense.
FFR and endothelial dysfunction
The endothelium plays an important role in the regulation of myocardial flow.18 Increased shear stress induces vasodilatation in normal arteries. Dysfunctional endothelium is a hallmark of atherosclerosis and it can be postulated that the vast majority of patients in whom FFR is measured have some degree of endothelial dysfunction. This has two practical consequences. (1) All measurements of FFR should be performed after the administration of nitrates, an endothelium independent vasodilator, to minimise the risk of paradoxical vasoconstriction. All FFR validation studies have been performed after administration of nitrates. (2) Severe endothelial dysfunction with paradoxical vasoconstriction during exercise might be responsible for rare cases of exercise induced ischaemia despite an FFR >0.80 as obtained after nitrates in the catheterisation laboratory.
You can get CPD/CME credits for Education in Heart
Education in Heart articles are accredited by both the UK Royal College of Physicians (London) and the European Board for Accreditation in Cardiology—you need to answer the accompanying multiple choice questions (MCQs). To access the questions, click on BMJ Learning: Take this module on BMJ Learning from the content box at the top right and bottom left of the online article. For more information please go to: http://heart.bmj.com/misc/education.dtl
RCP credits: Log your activity in your CPD diary online (http://www.rcplondon.ac.uk/members/CPDdiary/index.asp)—pass mark is 80%.
EBAC credits: Print out and retain the BMJ Learning certificate once you have completed the MCQs—pass mark is 60%. EBAC/ EACCME Credits can now be converted to AMA PRA Category 1 CME Credits and are recognised by all National Accreditation Authorities in Europe (http://www.ebac-cme.org/newsite/?hit = men02).
Please note: The MCQs are hosted on BMJ Learning—the best available learning website for medical professionals from the BMJ Group. If prompted, subscribers must sign into Heart with their journal’s username and password. All users must also complete a one-time registration on BMJ Learning and subsequently log in (with a BMJ Learning username and password) on every visit.
CONCLUSIONS: TOWARDS A NEW DIAGNOSTIC ALGORITHM FOR PATIENTS WITH KNOWN OR SUSPECTED CORONARY ARTERY DISEASE
Pressure derived FFR is a theoretically robust and practically simple means of assessing the functional consequences of epicardial coronary atherosclerosis. With minimal experience the technique of FFR measurements is simple, swift and safe. Only a pressure wire and a bolus of hyperaemia inducing medication are required. Its invasive nature is counterbalanced by its unequalled spatial resolution, offering functional information down to the “per centimetre” level, while non-invasive tests operate, at best, at the “per vascular territory” level. Clinical outcome data of patients in whom the revascularisation strategy has been based on FFR measurements are very encouraging. Accordingly, FFR can be considered as the interventional cardiologist’s “pocket myocardial perfusion imaging” modality. This is true with some important qualifications: (1) FFR is more accurate in intermediate lesions; (2) FFR has a better spatial resolution; (3) combined with the index of myocardial resistance (IMR), FFR is able to distinguish epicardial and myocardial resistance19 (this aspect is beyond the scope of this review); and (4) it is available in the catheterisation laboratory, as FFR is performed in conjunction with coronary angiography. Therefore, it is the only true “all-in-one” approach for patients with suspected or known coronary artery disease as it combines unequalled physiological information, the best possible anatomical information, and the possibility of immediate revascularisation if needed.
Competing interests: In compliance with EBAC/EACCME guidelines, all authors participating in Education in Heart have disclosed potential conflicts of interest that might cause a bias in the article. The authors have no competing interests.
Funding: This work was supported by the Meijer Lavino Cardiac Research Foundation.
▸ Additional references are published online only at http://heart.bmj.com/content/vol94/issue7
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.