The heart is the engine that powers life, and the coronary circulation represents the fuel pipes, providing blood with oxygen and other nutrients to keep the heart beating. Under normal physiologic circumstances, there will always be an equilibrium between oxygen demand of the myocardium and oxygen supply provided by the blood flow in the coronary arteries. There is an ingenious regulation system in the coronary circulation to maintain this equilibrium, called autoregulation. Due to the enormous reserve of the coronary circulation to provide blood to the myocardium, early stages of coronary atherosclerosis and narrowing in the coronary arteries will hardly be noticed, and if the coronary arteries become more severely narrowed, complaints will only occur in situations where the oxygen demand is increased, such as physical exercise or stress. Under those circumstances, myocardial ischemia will present itself by a characteristic pain or unpleasant sensation in the chest, arms, neck, or back, called angina pectoris. However, under resting circumstances, blood flow in the coronary circulation can be kept sufficient for a long time despite the presence of important narrowing.1
For those reasons, it is important to realize that in assessing the coronary circulation, it is not the coronary blood flow at rest that should be studied as a measure for the severity of coronary artery disease, but the maximal achievable blood flow as can be provoked by maximum exercise or vasodilatory stimuli like adenosine. Stated in another way, the degree to which maximum blood flow in the coronary circulation is decreased determines the exercise level at which angina pectoris occurs.
Also, in case of acute coronary syndromes, fundamentally the same principles hold true. However, in those situations, a rapid and sharp decrease of coronary blood flow usually occurs due to plaque rupture and/or thrombosis, often superimposed upon preexisting lesions.
It should be realized that the severity of disease on the coronary angiogram has only a poor correlation with the degree to which coronary blood flow is decreased. In other words, it is difficult to assess the physiologic significance of a coronary artery stenosis from the coronary angiogram.2,3 Nevertheless, the angiogram is important because in performing coronary interventions, it is a road map for the interventionalist to manipulate with wires and other equipment and to place the stent(s). In this chapter, the structure of the coronary circulation is discussed, followed by the regulation of coronary blood flow. There is a brief discussion on the development of atherosclerosis and the physiologic methods to detect this in early and later stages. Some words are also spent on coronary stenoses and myocardial ischemia and the consequences of ischemia for the heart. This is followed by the paramount question why functional testing is important, leading to the concept of fractional flow reserve. Some physiologic aspects on background and features of fractional flow reserve are briefly discussed. The clinical use of fractional flow reserve is discussed later in Chapter 24. Finally, some specific manifestations of coronary artery disease and physiologic methods for how to address these are discussed.
In most mammalians, the coronary circulation consists of 2 blood vessels originating from the proximal aorta. In humans, these coronary arteries have a diameter of approximately 2.5 to 4.0 mm and run as a kind of corona around and across the heart. Just like all other arteries, these coronary arteries have a layered structure with an intima, media, and adventitia. It is important to realize that under normal circumstances, the resistance of the coronary artery to blood flow is negligible, even at maximum hyperemia. From the epicardial coronary arteries, perforating branches enter the myocardium, and within the myocardium, these divide further into smaller vessels, called arterioles, with a diameter between 100 and 400 μm. Around these arterioles, smooth muscle cells are located in groups and are able to vary the resistance of the arterioles over a large domain (Fig. 6-1).
Schematic representation of the coronary circulation. The coronary artery splits up into arterioles with a diameter of 100 to 400 μm, which further divide into the capillary bed. At the entrance of the arterioles, muscular sphincters are located that can vary the resistance of the arterioles by at least 500%. Under resting conditions, the sphincters will be constricted, and if a higher blood flow is needed in response to higher oxygen demand (as in exercise), the sphincters dilate. In that way, at an equal perfusion pressure, myocardial blood flow can increase by at least 500% in healthy young individuals. In case of a moderate stenosis, an additional epicardial resistance occurs, and as a compensation, the arteriolar sphincter partially dilates even under resting conditions in order to maintain a total resistance in the bed equal to R. Consequently, during resting conditions, no noticeable effect of the epicardial stenosis is observable. However, during situations of increased demand (eg, exercise) the compensatory vasodilator reserve of the sphincters is already decreased and the maximal achievable blood flow in the coronary artery and myocardial bed is decreased accordingly. Therefore, myocardial ischemia and angina pectoris may occur during exercise. In case of a very severe stenosis in the epicardial coronary artery (lower pictograms), the epicardial stenosis can be so tight that the arteriolar sphincter needs to dilate completely to compensate the additional resistance in the coronary artery. This means that no reserve at all is left anymore. In such a case, even minimal increase in oxygen demand will result in angina pectoris. With further increase of stenosis severity, frank ischemia at rest will occur, resulting in unstable angina or an acute coronary syndrome. (From Pijls NHJ. Coronaire fysiologie en myocardischemie. In Van der Wall EE, Van de Werf F, Zijlstra F. Cardiologie. Houten, the Netherlands: Bohn Stafleu van Loghum, 2008, with permission of Springer.22)
For the purpose of simplicity, these smooth muscle cells around the arterioles are indicated as a kind of sphincter. In reality, the muscle cells are more equally distributed along the arterioles. The resistance of these arterioles can vary in healthy individuals by at least 500%, which means that under normal circumstances, maximum blood flow can be at least 5 times baseline blood flow. This ratio between maximum blood flow and baseline blood flow is called coronary flow reserve (CFR), as will be discussed more extensively later. The arterioles split up into the capillaries, in analogy to any other organ in the human body. The capillary network of the heart is very dense, and every myocyte is next to a capillary. Finally, the capillaries collect into small venules and collect further into veins, which either drain directly into the heart chambers (Thebesian veins) or collect in 1 large vein, the coronary sinus, with a diameter of 5 to 10 mm and ending in the right atrium.
Coronary blood flow equals approximately 3% to 5% of total cardiac output and varies from 200 mL at rest to approximately 1 L per minute at maximum exercise in healthy young people. It is important to realize that in contrast to most other organs in the human body, oxygen extraction from the blood in the coronary circulation is very high and that oxygen saturation in the coronary sinus is relatively low (30%-40%). This means that in the coronary circulation increase in oxygen consumption by the myocardium can only be achieved by an increase of blood flow and not by further extraction of oxygen. This issue will be discussed later in this chapter and in the other chapters.
Under baseline circumstances, the oxygen consumption of the myocardium is 0.7 to 1.3 mL O2/min/g of tissue, and under maximum hyperemic circumstances, this increases to 3 to 6 mL O2/min/g.
During exercise, heart rate and contractility will largely increase and lead to an increase in oxygen consumption; in addition, afterload will increase, thereby further increasing the oxygen demand and consumption of the heart. To match this tremendously increased demand, a wonderful regulation mechanism has been designed by nature, called autoregulation, which is further explained in Figure 6-1.
Collaterals are small blood vessels that can connect perfusion territories of different coronary arteries. The diameter of collaterals can vary from 10 μm to >1 mm. In the latter case, the collaterals are visible on the coronary angiogram. Collaterals can either be preexistent or develop as a response to myocardial ischemia. Repetitive ischemia is a necessary condition and the most potent stimulus to develop collaterals.
Collaterals can be either subendocardial or subepicardial. In the first case, the process underlying collateral development is often angiogenesis, and such angiogenetic collaterals only exist in a 1-layered rudimentary vessel structure. Subendocardial collaterals are often nicely observed running through the septum and connecting the septal branches of the left anterior descending artery and the posterior descending branch of the right coronary artery. In the case of subepicardial collaterals, these are often preformed (with a large variation between different individuals), have a 3-layer structure like a normal small artery, and are often very tortuous (corkscrew). This type of collaterals can often be seen across the apex connecting the distal right coronary artery and the left anterior descending artery, or running from the right ventricular branch of the right coronary artery to the left coronary artery, or even running across the atrium.
Collaterals are important because in a number of patients collaterals provide sufficient blood flow to maintain oxygen supply at rest even in case of a severe stenosis or total occlusion of the recipient coronary artery belonging to the collateral-dependent territory. The extent of collaterals and the intrinsic aptitude to develop them largely vary from one person to another and are also species dependent. In humans, in case of repetitive ischemia of the myocardium, collaterals may develop and reach a kind of maximum in most patients after approximately 3 months.
REGULATION OF CORONARY BLOOD FLOW
Autoregulation is the mechanism by which oxygen supply and demand of the myocardium are matched to each other both under physiologic and pathologic circumstances (see Fig. 6-1). It is important to realize that the arteriolar sphincters can vary the resistance over a large range.4 If these sphincters are completely dilated, resistance in the coronary circulation decreases by roughly 500%, which means that at an equal perfusion pressure, the reserve of coronary blood flow is also 500%. Under normal circumstances, the perfusion pressure in the coronary circulation is the difference between the aortic pressure and the central venous pressure (in normal circumstances, 0-5 mm Hg). If no coronary stenosis is present, the pressure in the distal coronary artery at the entrance of the arterioles equals the aortic pressure because there is no decline of pressure in a normal coronary artery, not even during maximum hyperemia. In case of increased oxygen demand of the myocardium (exercise, stress), the arteriolar sphincters can dilate, and blood flow can increase accordingly, even if perfusion pressure remains unchanged. The mechanism by which sphincters can dilate or contract is dependent, on one hand, on mechanical factors (eg, shear stress) and, on the other hand, on humoral factors (eg, bradykinin, acetylcholine, nitric oxide). If, at exercise, arterial blood pressure further increases (and thereby also the perfusion pressure across the coronary circulation), coronary blood flow will be able to increase further and will always remain sufficient to match the increased oxygen demand and metabolic needs of the myocardium.
The coronary autoregulation also keeps coronary flow constant despite changes in blood pressure. If, for whatever reason, blood pressure decreases, this is also accompanied by dilatation of coronary sphincters to keep coronary perfusion constant despite low blood pressure. If blood pressure increases inadvertently, contraction of sphincters prevents coronary blood flow from increasing inadvertently. In this way, coronary blood flow remains constant between mean aortic pressure from 50 to 130 mm Hg. At average blood pressures of greater than 130 mm Hg, sphincters are maximally contracted, and the further increase of blood pressure will increase coronary perfusion.
Below a mean arterial pressure of 50 mm Hg, the sphincters are maximally dilated, and a further decrease of blood pressure will lead to a pathologic decrease of coronary perfusion. As mentioned, the phenomenon whereby coronary perfusion can be maintained constant over a large range of perfusion pressures is called autoregulation of coronary blood flow.4
Under baseline circumstances, blood flow in the left coronary artery occurs mainly during diastole. In systole, the myocardium compresses its own vasculature, and there is almost no forward blood flow in the left coronary artery. This means that blood flow in the coronary arteries is antiphasic compared to other parts of the human circulation, where blood flow mainly occurs during systole.
During maximum hyperemia of the myocardium (as present during maximum exercise but also after administration of specific vasodilatory drugs), diastolic blood flow in the coronary artery further increases, but also a systolic component in coronary blood flow occurs consisting of 15% to 25% of total coronary blood flow (Fig. 6-2).
A and B. Blood flow pattern in a normal left coronary artery at rest and at hyperemia. At rest, systolic blood flow is very low, and flow mainly occurs during diastole. During maximum hyperemia, a systolic component also occurs, whereas diastolic blood flow increases even more. C. Example of reactive hyperemia after 20 seconds of occlusion of the left anterior descending artery of a dog. After relief of the occlusion, a rapid increase in blood flow occurs to 500% of resting blood flow, followed by a decrease when the oxygen debt is solved. D. Reactive hyperemia after intracoronary administration of 10 mg of papaverine in the same dog. E. Phasic and mean coronary blood flow in the left anterior descending artery of a dog after occlusions of 3, 5, 10, 20, and 30 seconds. As the time of occlusion increases, reactive hyperemia is more pronounced up to an occlusion time of approximately 20 seconds. With longer occlusions, the peak hyperemia no longer increases, but it takes longer before flow returns to baseline. (From Pijls NHJ. Coronaire fysiologie en myocardischemie. In Van der Wall EE, Van de Werf F, Zijlstra F. Cardiologie. Houten, the Netherlands: Bohn Stafleu van Loghum, 2008, with permission of Springer.22)
The blood flow pattern in the right coronary artery is more variable and more equally distributed across systole and diastole. In case of a dominant right coronary artery, it can look like a flow pattern in the left coronary artery. Normal blood flow patterns at rest and hyperemia are indicated in Figure 6-2.
If, in an experimental model (eg, a dog or a pig), the coronary artery is occluded during some time, myocardial ischemia will occur, and after release of the occlusion, a temporary increase in flow to the myocardium will occur. This is called reactive hyperemia (see Fig. 6-2). The longer the occlusion, the more pronounced and longer is the reactive hyperemia. Reactive hyperemia is maximal after occlusions of at least 20 seconds. If the occlusion lasts longer, the peak of hyperemia will not further increase but recovery of flow to baseline will take longer. Obviously, after occlusions of 20 seconds or more, the arteriolar sphincters are completely dilated. Maximum coronary hyperemia can not only be obtained by temporary occlusion of the coronary artery, but also pharmacologically by administration of specific drugs such as intracoronary papaverine or adenosine or intravenous adenosine, adenosine triphosphate, or regadenoson.5,6
The level to which coronary or myocardial blood flow can increase is indicated by CFR. The absolute CFR is defined as the ratio between hyperemic and baseline blood flow (see Fig. 6-2; Fig. 6-3).7
Definition and characteristics of coronary flow reserve (CFR) and fractional flow reserve (FFR). In both panels, the relationship between coronary perfusion pressure and coronary blood flow is indicated. Between an average perfusion pressure of 50 and 130 mm Hg, coronary perfusion remains constant at rest (autoregulation, horizontal lines). At maximum exercise or after administration of a vasodilatory stimulus, the arteriolar sphincters in the coronary artery are fully dilated, and a linear relation exist between perfusion pressure and blood flow (oblique lines). In case of a coronary stenosis, the slope of the oblique line will decrease. CFR is defined as maximum blood flow divided by blood flow at rest. Maximum blood flow is dependent on blood pressure, and blood flow at rest varies with heart rate and contractility. Consequently, CFR can vary over a large range for 1 specific coronary stenosis. FFR is defined as maximum blood flow in the presence of a stenosis divided by maximum blood flow in the hypothetical case where no stenosis would be present at all. This ratio is independent of changes in blood pressure, heart rate, and contractility, and under normal circumstances, FFR always has the same reference value (ie, 1.0). An FFR below 0.75 to 0.80 generally corresponds with inducible myocardial ischemia of the territory supplied by that particular coronary artery. In such cases, the stenosis is considered hemodynamically significant, and revascularization is most likely indicated.
As indicated in Figure 6-3 and as is clear from theoretical considerations, CFR correlates with blood pressure and is higher with higher blood pressure. CFR is also dependent on age and decreases with aging. In healthy young people, normal CFR is 5 to 6, but in healthy octogenarians, CFR is often hardly 1.5.
CFR is a useful theoretical parameter to understand the coronary circulation but less suitable in clinical practice to quantify the hemodynamic significance of a coronary stenosis because of its dependency on blood pressure, age, and a number of other factors. In the practice of the interventional laboratory, apart from the lack of a normal reference value, probably the most important limitation is that to determine CFR reliably, one has to rely on resting blood flow. In the catheterization laboratory, in a patient laying on the table and undergoing an intervention, it is hard to be sure that true resting blood flow is present. As a result of these limitations of CFR in clinical studies, the threshold of CFR below which coronary ischemia is inducible varies in different studies from 1.6 to 3.3. This means that the same CFR value of, for example, 2.5 can be completely normal in one person but severely decreased in another person. Consequently, it is difficult to use CFR for decision making with respect to revascularization.
To overcome these shortcomings of CFR for clinical decision making, the concept of fractional flow reserve (FFR) has been developed.8 FFR is defined as maximum achievable blood flow in a stenotic coronary artery compared to maximum blood flow in that same coronary artery in the hypothetical case that the vessel would be normal. As will be explained later, FFR can be determined in the catheterization laboratory by coronary pressure measurement and is useful to characterize the hemodynamic significance of a coronary stenosis and to decide upon sense or nonsense of mechanical revascularization (stenting of bypass surgery). The concept of FFR is illustrated in Figure 6-3 and Figure 6-4 and discussed further later in this chapter.
Definition of fractional flow reserve (FFR). On the left side, the epicardial coronary artery and the myocardial vascular bed are depicted. In the upper part, no stenosis is present, and normal perfusion pressure during hyperemia is 100 mm Hg. In the lower part, a stenosis is present, and distal coronary pressure at hyperemia has decreased to 70 mm Hg. Therefore, the perfusion pressure has decreased to 70 mm Hg only, whereas it should normally be 100 mm Hg. Because of the linear relationship between hyperemic perfusion pressure and hyperemic blood flow (right part of the figure), this means that the maximum myocardial perfusion in the presence of a stenosis has decreased to only 70% of what would be its normal value. In fact, a ratio of hyperemic blood flows is expressed as a ratio of perfusion pressures, and FFR is the hyperemic distal perfusion pressure (Pd) in the presence of the stenosis divided by the perfusion pressure in the normal case, indicated by Pa. FFR indicates the fraction of normal maximum blood flow that is still achievable in the presence of a stenosis. An FFR of 0.7 means that, as a consequence of the epicardial coronary stenosis, maximum blood flow to the supplied myocardial territory has decreased to 70% of the value it should be in a normal case. Therefore, FFR is a specific measure of the functional severity of a coronary stenosis.
ENDOTHELIUM AND DEVELOPMENT OF ATHEROSCLEROSIS
The early phase of atherosclerotic coronary disease is endothelial dysfunction. This is invisible by any imaging method but can be demonstrated by functional testing (Fig. 6-5). In reaction to a number of humoral or mechanical stimuli (eg, increased shear stress), endothelial cells produce nitric oxide (NO), leading to relaxation of underlying smooth muscle cells and vasodilation with increase of blood flow in the coronary artery. The earliest stage of (coronary) atherosclerosis is endothelial dysfunction. This can be demonstrated by intracoronary acetylcholine administration. In the presence of healthy endothelium, acetylcholine also increases NO production and, thereby, coronary blood flow. Acetylcholine also has a direct vasoconstrictive effect on the underlying smooth muscle cells, but in healthy persons, this effect is weaker than the vasodilatory NO-stimulating effect. If the endothelium is diseased, the direct vasoconstrictive effect predominates, and so-called paradoxical vasoconstriction occurs. This phenomenon can be used in the interventional laboratory to demonstrate endothelial dysfunction. Such a test is performed by administrating subselectively intracoronary acetylcholine under controlled conditions but is not trivial and not without risk, and therefore, it should only be applied in specialized catheterization laboratories.
Abnormal endothelial function. The earliest stage of atherosclerosis is endothelial dysfunction, often present long before abnormalities on the angiogram can be seen. In the left panel, an apparently normal coronary angiogram is seen in a young male with many risk factors and a positive exercise test. To detect abnormal endothelial function, acetylcholine is injected into the coronary artery, resulting in severe and varying vasoconstriction in the different branches of the circumflex system. In the presence of normal endothelium, no such narrowing should be seen. In the right panel, nitroglycerin has been administered to relieve the acetylcholine effects.
The next step in the evolution of atherosclerosis is development of early plaques and often diffuse disease that is not visible on the angiogram but can be demonstrated by anatomic methods, such as intravascular ultrasound or optical coherence tomography, and by physiologic methods, such as FFR measurement. In the latter case, during hyperemia, an abnormal pressure decline will occur even in an apparently normal coronary artery.
Finally, when macroscopic, gross atherosclerosis occurs, it can be detected easily on the coronary angiogram, but physiologic measurements remain generally necessary to define the extent and physiologic significance of disease.
For the interventional cardiologist, this is important because in case of diffuse decline of coronary pressure along a coronary artery and a focal decline at 1 particular spot or a severe stenosis, stenting of the severe stenosis will not result in a restoration to normal because the diffuse gradient will persist within the artery and even increase after stenting the focal stenosis. It is important to quantify this because, in some cases, this diffuse disease as detected by the pressure pullback recording can be responsible for residual inducible ischemia, even after successful stenting of focal spots. Similarly, by such hyperemic pressure pullback recording, patients with diffuse disease can be distinguished who cannot be treated with mechanical techniques of revascularization even though ischemia may be present. This issue will be further discussed in Chapter 24.
CORONARY STENOSIS AND MYOCARDIAL ISCHEMIA
Atherosclerosis of the coronary arteries is common in the Western world and increases with age. Approximately half of asymptomatic and apparently healthy persons between ages 50 and 65 years have visible atherosclerotic abnormalities in the coronary arteries. It is important to note that, in many of these patients, such narrowing will never cause significant problems. Therefore, the anatomic presence of a coronary stenosis does not indicate per se that revascularization is indicated. However, in case of visible atherosclerosis, preventive lifestyle management and medical treatment using aspirin, statins, angiotensin-converting enzyme inhibitors, or β-blockers might be indicated and useful.
In case of a developing stenosis in the coronary artery, an additional resistance in the coronary circulation occurs. At rest, this can be easily compensated by compensatory dilatation of the arteriolar sphincters, as can be easily understood from Figure 6-1. If at baseline, sphincter resistance is called R, minimal resistance (in this example) will be 1/5 R. A moderate stenosis in the supplying coronary artery with a resistance equaling 2/5 R can be easily compensated by a decrease of arteriolar resistance to 3/5 R.
In other words, there are now 2 resistances in series, but at rest, the total resistance remains constant assuming unchanged perfusion pressure. Therefore, coronary flow at rest will also remain constant. However, this is at the cost of CFR, which is decreased because at maximum exercise the remaining vasodilatory capacity of the sphincters has been decreased.
If the coronary narrowing becomes more severe, the situation can occur that the full vasodilatory capacity of the distal sphincters is mandatory to compensate the increased resistance in the coronary artery and no further reserve is left.1 As will be clear, this is the situation with stable angina pectoris, which increases over time. With further increase of stenosis severity, even ischemia at rest may occur. Acute coronary syndromes often occur from a sudden increase of resistance in the coronary artery by plaque complication, rupture, or thrombosis. Although it is often believed that plaque rupture occurs in previously minor plaques, recent studies have indicated that the hemodynamic significance of such underlying plaques is often significant and the resultant repetitive ischemia is a potent stimulus to make a plaque vulnerable. In any case, there is a strong relationship between repetitive ischemia and vulnerability.9,10
Whatever the relationship between vulnerability and ischemia may be, it is clear that inducible ischemia of the myocardium is the most important prognostic factor for patients with coronary artery disease. If a coronary stenosis causes no inducible ischemia, its prognosis is generally favorable, and medical treatment is the best way to continue.11 If a coronary stenosis is hemodynamically significant, outcome in terms of prognosis is much worse and a coronary intervention (either stenting or bypass surgery) is generally indicated.11,12,13 In addition, hemodynamically significant stenoses often cause complaints (angina pectoris), and mechanical intervention is the most efficient way to treat such complaints.
Consequently, one of the most important issues in interventional cardiology is to make correct decisions about which coronary stenoses should and should not be revascularized. Beyond doubt, the functional significance of a stenosis is much more important in this respect than its appearance on the angiogram. The gold standard used in the catheterization laboratory to assess the hemodynamic significance of a coronary stenosis is FFR. The physiologic background of FFR and some specific features of it will be discussed in the next section, whereas a more extensive discussion about its practicalities and clinical application will be outlined in Chapter 24.
The concept of FFR is based on 2 important principles. First, it is not resting blood flow but maximal achievable blood flow that determines the functional capacity of a patient and is decisive regarding whether myocardium will become ischemic. Second, at maximum vasodilatation (corresponding with maximum hyperemia or maximum exercise), blood flow to the myocardium is proportional to myocardial perfusion pressure, as outlined earlier.
With those principles in mind, FFR is easy to understand from Figure 6-4. In the left upper part of the figure, a normal coronary artery and its myocardial perfusion territory are represented. If maximum vasodilation in the coronary circulation is present (all vessels and sphincters dilated), blood flow will be proportional to aortic pressure (Pa) minus central venous pressure (Pv), which is put at 0 for case of simplicity. In the left lower part of the figure, a coronary stenosis is present, resulting in a particular pressure drop at hyperemia within the coronary artery. To understand the functional significance of that stenosis for the patient, it is not the gradient across that stenosis that is important, but the degree to which the distal perfusion pressure has decreased. In our example, perfusion pressure has decreased to 70 mm Hg, whereas it should be 100 mm Hg in a normal case. Because of the proportionality between perfusion pressure and coronary blood flow at maximum hyperemia (assuming that minimal resistance is constant), maximum achievable blood flow in the diseased situation has decreased to 70% of its normal value (right part of the figure). In other words, a ratio of maximum blood flows is expressed as a ratio of perfusion pressures. In contrast to flow (which is difficult to measure directly in human coronary arteries), pressure can be measured easily by a 0.014-inch pressure wire. For that reason, FFR can be easily assessed in the practice of the catheterization laboratory by a pressure wire that measures distal coronary pressure (Pd) after administration of a maximum hyperemic stimulus and by comparing this distal coronary pressure at hyperemia to aortic pressure. Or simply stated:
FFR = (Pd – Pv)/(Pa – Pv) ≈ Pd/Pa
In clinical practice, measurement of FFR is easy, as illustrated in Figure 6-6. In fact, the concept of FFR encompasses much more than only the effect of a coronary stenosis on myocardial blood flow. It is also possible to express maximum coronary artery blood flow, myocardial blood flow, and collateral blood flow quantitatively as a percentage of normal maximum myocardial blood flow. To measure the separate contributions of coronary artery and collateral blood flow to myocardial flow, however, knowledge of coronary wedge pressure is mandatory (as can be measured during balloon occlusion). In fact, FFR gives a complete description of the coronary circulation and all components of it in terms of pressure measurement. In-depth discussion about the concept of FFR is beyond the scope of this chapter, and for that purpose, we refer the reader to the initial publication defining FFR in Circulation in 19938 and a myriad of review publications.14,15,16
A. Measurement of fractional flow reserve (FFR) in clinical practice in a patient with angina pectoris, a positive exercise test, and a suspected lesion in the proximal left anterior descending artery (arrow). B. A pressure wire is introduced into the coronary artery. Aortic pressure measured by the guiding catheter is indicated by the red signal, and the coronary pressure measured by the pressure sensor is indicated by the green signal. When the sensor is proximal to the suspected lesion, pressures are equal. When the pressure sensor crosses the lesion, a sudden drop in distal pressure occurs, and when inducing maximum hyperemia by IV administration of adenosine 140 μg/kg/min, distal pressure further decreases, reflecting increase of blood flow. C. At steady-state maximum hyperemia, distal coronary pressure is 52 mm Hg compared to 91 mm Hg in the aorta. FFR is easily calculated as 52/91 = 0.57. In other words, as a consequence of the visible plaque in the proximal LAD on the angiogram, maximal achievable blood flow to the anterior wall of the heart in this patient has decreased to only 57% of its normal value.
PHYSIOLOGIC FEATURES OF FRACTIONAL FLOW RESERVE
FFR has a number of specific advantages over other physiologic indexes that make it an easy index to assess coronary artery disease and to make decisions about the necessity of interventions. These physiologic features will be discussed in the following section. For more practical discussion, see Chapter 24.
Uniform Normal Value and Discrimination of Inducible Ischemia
As discussed earlier, in a completely normal coronary artery, there is no decline of pressure, not even during maximum hyperemia.17 Consequently, the normal reference value of FFR is 1.0 irrespective of the patient, hemodynamic conditions, age, or any other variable. Having an uniform normal value is important, because without that, it will be impossible to define a reliable threshold value for ischemia. The threshold value of ischemia for FFR is generally taken as 0.80. In fact, it is somewhere between 0.75 and 0.80 with a small gray zone in between, as will be discussed later. In clinical practice, it is generally felt justified to revascularize a stenotic coronary artery if FFR is ≤0.80 and to give medical treatment if FFR is >0.80.
Relationship Between Vessel Size and Perfusion Territory
In contrast to any anatomic method, FFR accounts for the relationship between the severity of the epicardial coronary stenosis and the size of the perfusion territory. This is especially important in assessing stenoses severity after previous myocardial infarction. Two coronary arteries might have an exactly identical stenosis with exactly identical morphologic features, but the functional severity will vary if the perfusion territory has a different extent. Such difference cannot be detected when only looking at the morphologic features of the stenotic artery. In the case of 2 identical stenoses but a larger (viable) perfusion territory distal to 1 of these lesions, the hyperemic response will be larger and the FFR consequently lower (Fig. 6-7). This explains why, after previous myocardial infarction, an apparently severe stenosis might have a rather high FFR, whereas a similar stenosis without infarction has a much lower FFR. It also explains why a moderate lesion with a large perfusion territory, including collaterally dependent myocardium, might be significant, whereas it would not be significant without the collateral perfused territory.
Fractional flow reserve (FFR) is not only a measure of the functional severity of the coronary artery stenosis itself, but also relates stenosis severity to the extent of the perfusion territory, as indicated in this figure. In both the upper and lower parts of this figure, a similar stenosis is present, and with anatomic methods, no difference can be observed. However, because in the upper part of the figure the perfusion territory is larger, a higher hyperemic response can be provoked and FFR will be lower than in the case of a decreased perfusion territory. In fact, FFR connects coronary stenosis severity, coronary blood flow, inducible ischemia, and extent of perfusion territory with each other. See text for further explanation.
Independency from Blood Pressure, Heart Rate, and Contractility
In contrast to CFR, FFR is not dependent on changes in hemodynamics such as heart rate, blood pressure, and contractility.18 This is due to the fact that FFR is not dependent on resting blood flow, as can be understood from Figure 6-3. True resting conditions in the catheterization laboratory are difficult to obtain, and normally, during cardiac catheterization and intervention, large variations in so-called resting blood flow occur. This explains in part the large fluctuation in CFR (and other indexes relying on resting blood flow) and the wide range of CFR values in different patients with the same FFR. As illustrated in Figure 6-3, such limitation is not present for FFR. Also, in humans, studies have been performed with large variations in heart rate, blood pressure, and contractility with no effect on FFR.18
In the literature, it is often suggested that different values of CFR in different patients with the same FFR is a measure of microvascular dysfunction. However, it is likely that such a range of CFRs is due to absence of one uniform normal value (eg, age, different hemodynamic circumstances) and has little to do with abnormal microvascular status.
As mentioned earlier, FFR accounts for the effects of the collateral circulation. In fact, the separate contribution of collateral and coronary artery blood flow to myocardial perfusion can be calculated by the pressure measurements, provided that wedge pressure is known.8
In case of extensive collaterals, a stenosis might have a high FFR value, whereas in the absence of collaterals, the FFR might be low. Conversely, as mentioned earlier, a similar stenosis may have a lower FFR value if the respective artery gives collaterals to another territory.
Issue of Maximum Hyperemia
As discussed earlier, one of the fundamental requirements to calculate FFR is the presence of maximum coronary and myocardial hyperemia. In the catheterization laboratory, this can be achieved pharmacologically in a number of ways, as summarized in Table 6-1. When using maximum hyperemia, FFR is an extremely reliable index of ischemia and has been validated versus a true gold standard using a so-called prospective multitesting Bayesian approach.19
Table 6-1Effects of Different Vasodilatory Stimuli on the Epicardial Coronary Artery and the Microvascular Bed ||Download (.pdf) Table 6-1 Effects of Different Vasodilatory Stimuli on the Epicardial Coronary Artery and the Microvascular Bed
Although a number of ways for inducing hyperemia are available in the catheterization laboratory and not difficult with a good logistic setup, because of simplicity, it has been proposed recently to leave hyperemia out and to perform resting measurements only. Several resting indexes have been proposed, such as the Pd/Pa at rest and the so-called iFR (instantaneous wave-free ratio).20 Unfortunately, the physiologic basis for using resting indexes is poor and not supported by experimental studies in animals. Generally, according to Poiseuille’s law and as validated in animals and humans, a small baseline pressure gradient across the stenosis can be accompanied by a large hyperemic gradient, especially in young patients, patients with a short severe stenosis, and patients with proximal lesions or a large perfusion territory, whereas a moderate gradient with only moderate hyperemic increase can be observed in older patients, patients with distal lesions, and patients with small coronary arteries (Figs. 6-8 and 6-9).
Explanation of why resting pressure indexes are not sufficient to predict true stenosis severity. According to Poiseuille’s law, the pressure gradient across a stenosis is related to blood flow by a linear and quadratic component, representing friction (f) and separation (s) coefficients, respectively. In some lesions, friction is the predominant factor, which means that a moderate resting gradient will moderately increase at hyperemia. At the other end of the spectrum, where separation is predominant, a minimal resting gradient can be accompanied by a very large hyperemic gradient. The latter is especially the case in younger patients with short lesions in the proximal part of a large coronary artery with a large perfusion territory. See also Figure 6-9. LAD, left anterior descending artery.
Example of a young gentleman with a 70% short stenosis in a large dominant right coronary artery. At rest, no gradient is present at all. At maximum hyperemia, however, a large pressure gradient develops, and fractional flow reserve is 0.53, indicating severe ischemia in the inferior wall. The pressure recording also nicely illustrates the principle of a pressure pullback recording, enabling the exact location of a pressure drop and predicting the success of coronary stenting.
Therefore, in clinical practice, it is generally unpredictable to what extent a baseline gradient will increase with hyperemia. Nevertheless, if a resting gradient is large and already below the ischemic threshold of FFR, just for the mere purpose of making the decision about whether or not to revascularize, no additional hyperemia will be necessary, whereas in the case of very small resting gradients, often (but not always) no inducible ischemia will be present. In addition, as mentioned earlier, it is often not possible to achieve true resting conditions in a human catheterization laboratory. When using resting indexes in the best possible way, agreement with FFR with respect to presence of ischemia is achieved in approximately 80% of cases. FFR in itself has an accuracy of approximately 95% compared to the true gold standard.19,21
There is another reason why true hyperemia is desirable. In the current population of patients in the catheterization laboratory, who often have complex coronary artery disease, predicting whether a stenosis is causing ischemia is only 1 piece of information. The other piece is making a hyperemic pullback recording to assess abnormalities along the complete course of the coronary artery to understand the nature of the disease and to predict whether stenting will be effective. For such pullback recording, resting conditions lack sufficient spatial resolution (poor signal-to-noise ratio), and maximum hyperemia is mandatory.
A compromise between resting indexes and hyperemia might be achieved by contrast FFR (cFFR; defined as Pd/Pa after a single regular bolus of 10 mL of contrast agent). An injection of contrast agent into a coronary artery generally increases blood flow very shortly up to a level of at least 80% of maximum hyperemia. Therefore, it has been suggested to use a contrast injection to discriminate whether FFR is abnormal. Although suboptimal, cFFR is more reliable than using resting indexes only, and it is easy to obtain. In addition, lesions with a small resting gradient but large hyperemic gradient, which would be completely overlooked by resting indexes, will be easily discovered by contrast. Agreement between cFFR and true FFR is achieved in approximately 85% of patients (Fig. 6-10). Further investigation of these resting indexes and cFFR is mandatory.
The pyramid of diagnostic accuracy of functional indexes to correctly predict ischemia: Accuracy of the standard angiogram, resting pressure indexes, contrast FFR and true FFR to predict coronary ischemia and to facilitate decision making with respect to revascularization. For further explanation, see text. iFR, instantaneous wave-free ratio.
SOME SPECIFIC MANIFESTATIONS OF CORONARY ARTERY DISEASE
Diffuse Coronary and Microvascular Disease
When discussing CFR and FFR, for simplicity, it is often assumed that there is 1 (or several) focal narrowing in a coronary artery. However, atherosclerosis is a diffuse disease process, and often, abnormalities will be present along the complete course of the coronary artery, as can be noticed by intravascular ultrasound or optical coherence tomography. This diffuse disease can be so severe that focal stenting or bypass surgery does not make sense. Similarly, a lot of attention is paid to microvascular disease affecting the invisible part of the coronary angiogram. It might be clear that microvascular disease will affect both CFR and FFR. Several indexes have been developed to quantify microvascular disease (eg, index of microvascular resistance and absolute coronary blood flow). The practical consequences are 2-fold; first is the understanding that sometimes angina pectoris can occur without focal epicardial disease, and second is the necessity of aggressive medical treatment in such patients. However, the importance and prevalence of microvascular disease should not be overestimated; in the majority of patients with angina pectoris, inducible ischemia, and a focal lesion, stenting or bypass surgery results in disappearance of complaints and disappearance of inducible ischemia, indicating that the epicardial disease was the predominant factor.
When performing a coronary angiogram, sometimes bridging of the left arterial descending artery is observed during systole, due to a partly intramyocardial course of the artery or to a muscle bridge across the blood vessel. Bridging increases during adrenergic stimulation and may lead to complete occlusion during systole. However, its clinical meaning is mostly minimal. It is often unclear how this bridging should be interpreted and if it can be held responsible for complaints. In the literature and in clinical practice, many confounding points of view exist.
In our view, complaints due to bridging can only be present during adrenergic stimulation. This means that complaints at rest should never be ascribed to myocardial bridging, simply because coronary blood flow at rest is predominantly diastolic. As discussed earlier in this chapter, at maximum hyperemia, the systolic component in left coronary artery blood flow equals 15% to 25% of total flow (see Fig. 6-2). If that systolic component is 20% to 25%, one can imagine that by losing 20% to 25% of flow during occlusion of the bridge at maximum exercise, a theoretical FFR value of less than 0.75 to 0.80 occurs and could be responsible for ischemia. FFR in itself is not a good method to assess bridging because, at occlusion, there is no blood flow and consequently no gradient. Therefore, in those rare cases in which bridging is suspected to be responsible for myocardial ischemia, we use a Doppler wire to measure the coronary flow pattern in the left anterior descending artery. First, we induce maximum hyperemia using adenosine and calculate the systolic component of maximum coronary blood flow. Next, we perform flow velocity registrations at dobutamine-induced hyperemia. Only if the systolic component is more that 20% of the total velocity integral with adenosine and the systolic component disappears completely during dobutamine do we believe that it is justified to hold the bridging responsible for complaints. The approach to evaluate patients with myocardial bridging, as used in the Catharina Hospital (Eindhoven, The Netherlands), is outlined in Table 6-2.
Table 6-2Diagnostic Approach to Patients with Chest Pain and Myocardial Bridging of the Left Anterior Descending Artery in the Catharina Hospital ||Download (.pdf) Table 6-2 Diagnostic Approach to Patients with Chest Pain and Myocardial Bridging of the Left Anterior Descending Artery in the Catharina Hospital
|Step 1 || |
Complaints (also) present at rest? → not due to bridging.
Forget the bridge as cause of the complaints. No further tests in this respect.
If complaints only present during exercise → go to Step 2.
|Step 2 || |
Perform MIBI SPECT with maximum dobutamine* 40 μg/kg/min (not with dipyridamole/adenosine!).
Is there reversible defect in anterior wall/apex?
If yes → go to Step 3.
If no → bridge as cause of complaints extremely unlikely.
|Step 3 || |
Invasive intracoronary flow-velocity measurement with Doppler wire.
Make at first recording of systolic/diastolic velocity pattern with intravenous adenosine 140 μg/kg/min and thereafter with maximum dobutamine 40 μg/kg/min.b
If systolic component with adenosine is 20% of total flow velocity-time integral and completely disappears during dobutamine, transient ischemia during exercise due to bridging is likely.
CONCLUSIONS AND FINAL REMARKS
The coronary circulation is designed in such a way that there is always a match between oxygen demand and supply under a wide variety of physiologic conditions. This match is maintained by an ingenious regulation system called autoregulation.
Atherosclerosis of the coronary arteries is an insidious disease interfering with the structure and function of the coronary circulation. Disease is mostly present long before coronary artery abnormalities can be seen. In case of a visible stenosis in a coronary artery, there is only a poor relation between anatomic severity and the functional severity, and sophisticated physiologic examination may be necessary to establish whether ischemia is present and revascularization is warranted.
The gold standard for functional assessment in coronary artery disease is FFR, which is calculated during maximum coronary hyperemia by measuring distal coronary pressure using a pressure wire and comparing it to proximal pressure measured using the guiding catheter. FFR allows functional assessment in a practical and easy way and can be measured during invasive coronary angiography. The clinical use of FFR will be discussed further in Chapter 24.
C. Compensatory changes of the distal coronary vascular bed during progressive coronary constriction. Circulation
KL. Coronary anatomy to predict physiology: fundamental limits. Circ Cardiovasc Imaging
et al. Evolving concepts of angiogram: fractional flow reserve discordances in 4000 coronary stenoses. Eur Heart J
RJ. Regulation of coronary vasomotor tone under normal conditions and during acute myocardial hypoperfusion. Pharmacol Ther
et al. Intracoronary and intravenous adenosine
, and contrast medium to assess fractional flow reserve in humans. Circulation
et al. Single bolus intravenous regadenoson
injection versus central venous infusion of adenosine
for maximum coronary hyperaemia in fractional flow reserve measurement. EuroIntervention
M. Coronary flow reserve as a physiologic measure of stenosis severity. J Am Coll Cardiol
NH, van Son
et al. Experimental basis of determining maximum coronary, myocardial, and collateral blood flow by pressure measurements for assessing functional stenosis severity before and after percutaneous transluminal coronary angioplasty. Circulation
et al. A prospective natural-history study of coronary atherosclerosis. N Engl J Med
NH, van Schaardenburgh
et al. Percutaneous coronary intervention of functionally nonsignificant stenosis: 5-year follow-up of the DEFER Study. J Am Coll Cardiol
et al. Fractional flow reserve-guided PCI for stable coronary artery disease. N Engl J Med
et al. Fractional flow reserve versus angiography for guiding percutaneous coronary intervention in patients with multivessel coronary artery disease: 2-year follow-up of the FAME (Fractional Flow Reserve Versus Angiography for Multivessel Evaluation) study. J Am Coll Cardiol
NH. Fractional flow reserve to guide coronary revascularization. Circ J
JW. Functional measurement of coronary stenosis. J Am Coll Cardiol
WF. Percutaneous coronary intervention should be guided by fractional flow reserve measurement. Circulation
NH, Van Gelder
B, Van der Voort
et al. Fractional flow reserve. A useful index to evaluate the influence of an epicardial coronary stenosis on myocardial blood flow. Circulation
et al. Simultaneous coronary pressure and flow velocity measurements in humans. Feasibility, reproducibility, and hemodynamic dependence of coronary flow velocity reserve, hyperemic flow versus pressure slope index, and fractional flow reserve. Circulation
NH, De Bruyne
et al. Measurement of fractional flow reserve to assess the functional severity of coronary-artery stenoses. N Engl J Med
et al. Development and validation of a new adenosine-independent index of stenosis severity from coronary wave-intensity analysis: results of the ADVISE (ADenosine
Vasodilator Independent Stenosis Evaluation) study. J Am Coll Cardiol
C, van’t Veer
et al. VERIFY (VERification of Instantaneous Wave-Free Ratio and Fractional Flow Reserve for the Assessment of Coronary Artery Stenosis Severity in EverydaY Practice): a multicenter study in consecutive patients. J Am Coll Cardiol
Pijls NHJ. Coronaire fysiologie en myocardischemie. In Van der Wall
EE, Van de Werf
F. Cardiologie. Houten, the Netherlands: Bohn Stafleu van Loghum; 2008.
MULTIPLE CHOICE QUESTIONS
1. Which intracoronary parameter has been shown to be the least influenced by hemodynamic changes?
A. Coronary flow reserve (CFR)
B. Fractional flow reserve (FFR)
C. Instantaneous wave-free ratio (iFR)
D. Resting distal coronary pressure/aortic pressure (Pd/Pa)
2. In the normal heart at rest, an increase in coronary perfusion pressure of 20% will result in which of the following?
A. An increase in myocardial blood flow by about 20%
B. An increase in myocardial blood flow by about 40%
C. A decrease in myocardial blood flow by about 20%
D. No change in myocardial blood flow
3. In the normal heart during maximal hyperemia, an increase in coronary perfusion pressure of 20% will result in which of the following?
A. An increase in myocardial blood flow by about 20%
B. An increase in myocardial blood flow by about 40%
C. A decrease in myocardial blood flow by about 20%
D. No change in myocardial blood flow
4. What is the accuracy of the coronary angiogram to predict whether a coronary stenosis is able to induce myocardial ischemia?
5. What is the definition of fractional flow reserve (FFR)?
A. Pressure gradient over a stenosis during maximal hyperemia
B. Maximum blood flow over a stenosis divided by the resting blood flow over that stenosis
C. Maximum blood flow in the presence of a stenosis divided by the maximum blood flow in the hypothetical absence of that stenosis
D. Minimal luminal area divided by normal luminal area in case of a stenosis
Both animal and human studies have shown that FFR is the only parameter that is highly independent of hemodynamic variations such as heart rate, blood pressure, and contractility. The reason that FFR is fundamentally different than CFR, iFR, and resting Pd/Pa is that it is the only parameter that is not dependent on resting flow.
Coronary autoregulation keeps coronary flow constant despite changes in coronary perfusion pressure, between a mean aortic pressure of 50 and 130 mm Hg.
During maximal hyperemia (eg, induced by adenosine, severe ischemia, or a perfusion pressure out of the autoregulatory range), autoregulation is exhausted, and therefore, perfusion pressure is proportionally related to coronary flow.
The accuracy of the angiogram to predict functionally significant stenoses is rather poor—only 70%. Anatomy alone will never be sufficient to predict physiology because it does not incorporate several important determinants of maximal blood flow, including myocardial mass and microvascular function. FFR is the current gold standard in identifying hemodynamically significant stenoses and has an accuracy of about 95%.
During maximal hyperemia, autoregulation is exhausted and coronary perfusion pressure is proportional to coronary flow. In the normal vessel, the coronary pressure is the same throughout the vessel. Therefore, in a stenotic vessel, the proximal pressure is a reflection of what the distal pressure would have been in the absence of that stenosis. Consequently, by dividing distal pressure by the proximal pressure, one can determine the maximum blood flow in the presence of a stenosis divided by the maximum blood flow in the hypothetical absence of that stenosis.