When noninvasive testing for coronary artery disease (CAD) is inconclusive or suggests significant pathology, invasive testing is necessary. X-ray coronary angiography provides an overview of the coronary circulation and in particular helps to identify obstructive epicardial CAD. However, the coronary angiogram is often misleading. Significant-appearing CAD may not be responsible for myocardial ischemia and symptoms, whereas occult diffuse epicardial disease not apparent on the angiogram can be. Moreover, the coronary angiogram focuses on fixed obstructive epicardial disease, but it does not provide information regarding endothelial dysfunction or vasospasm, nor does it identify coronary microvascular dysfunction. A number of adjunctive techniques including both coronary wire-based measures and catheter-based systems allow for further interrogation of the coronary circulation at the time of coronary angiography. This chapter will focus on the main methods for assessing coronary physiology, namely coronary flow reserve, fractional flow reserve, and the index of microcirculatory resistance, as well as the two main methods for invasively imaging the epicardial coronary anatomy, namely intravascular ultrasound and optical coherence tomography.
Coronary Angiography
Coronary angiography is defined as the visualization of the coronary arteries after injection of contrast media. Typically, the angiographic diagnosis of CAD is made subjectively by the cardiologist who performed the procedure. In general, a visual estimation of the severity of a coronary narrowing is reported, for which a 50% stenosis is considered as obstructive coronary disease and greater than 70% stenosis as significant coronary disease. Classification systems aimed at standardizing the interpretation of an angiogram have been created that incorporate lesion characteristics such as degree of calcification, length, eccentricity, tortuosity, and location at a bifurcation. These techniques, however, are inherently limited by interobserver variability.
Quantitative coronary angiography is a computer-assisted method of measuring lesion length and stenosis severity. By using an object of known size, such as the catheter, to calibrate the system, quantitative coronary angiography ideally is less subjective and more accurate than other methods. Unfortunately, it too is prone to error and subjectivity due to operator technique. Despite these issues, the presence and severity of coronary disease as assessed by coronary angiography are predictors of long-term adverse outcome.
A number of limitations to angiography hamper its ability to accurately diagnose coronary disease, especially in the setting of moderate narrowing or diffuse disease. First, because the angiogram is a two-dimensional representation of a three-dimensional object, an eccentric narrowing can be missed if the correct angle is not used to image the vessel. Second, because a diseased area of a coronary artery is generally compared with an adjacent “normal” area, patients with diffuse disease without any focal component can be incorrectly classified as having normal coronary arteries. Finally, the angiogram highlights the lumen of the coronary artery but provides no information about the wall of the vessel. Positive remodeling of the artery at the site of atherosclerotic plaque development can result in preservation of the lumen and a near normal angiogram, which hide the atherosclerosis from the angiographer. Because of these limitations a number of adjunctive techniques have been developed to improve the invasive diagnosis of ischemic heart disease.
Coronary Flow Reserve
Coronary flow reserve (CFR) is defined as the ratio of the maximal or hyperemic flow down a coronary vessel to the resting flow. It can be measured invasively with a Doppler-tipped coronary guidewire that determines coronary velocity at rest and during hyperemia, typically induced with intracoronary or intravenous adenosine. Because velocity is proportional to flow, the coronary flow velocity reserve is a reflection of the CFR. If in addition to the velocity the area of the coronary vessel is known, the absolute CFR can be calculated. CFR also can be measured invasively by using a wire-based thermodilution technique. On one of the commercially available coronary pressure wires (St. Jude Medical, MIN), the pressure sensor also can act as a thermistor. With the commercially available software, the shaft of the wire acts as a proximal thermistor. Room temperature saline can be injected into the coronary artery and this system will calculate the transit time, which is inversely proportional to coronary flow. After three injections at rest, the resting mean transit time is calculated. Hyperemia is then induced with intravenous adenosine, and three injections are performed to determine the hyperemic mean transit time. CFR is measured in this situation by dividing the resting mean transit time by the hyperemic mean transit time. The thermodilution-derived CFR has been validated in animal and human models and has been compared in an animal model to a reference standard of absolute flow. It appears to correlate more closely to the standard than does Doppler-derived CFR.
A normal CFR is considered to be greater than 2.0 and in most patients should be somewhere between 3 and 5. Initially, invasive CFR was performed to interrogate the functional significance of an intermediate coronary stenosis with studies showing a correlation between CFR and noninvasive tests for ischemia. However, a number of limitations of invasively measured CFR impaired its broad clinical utility. First, it can be difficult to measure with a Doppler wire because of the challenge in obtaining a suitable Doppler signal. Second, because CFR relies on resting flow for its calculation, the repeatability of measurements is less than optimal. Any hemodynamic perturbation such as a change in heart rate, blood pressure, or left ventricular contractility will significantly change the CFR value as a result of the change in resting flow. The lack of a clear cut-off between a normal and abnormal CFR makes it difficult to use for clinical decisions. Because there is a range of normal CFR values between approximately 2.5 and 6, in one patient a value of 3.0 might be normal whereas in another patient normal CFR may be 5.0 and therefore a recorded value of 3.0 could be quite abnormal. Finally, by definition CFR is a measure of the entire coronary circulation. It interrogates the epicardial vessel as well as the coronary microvasculature ( Fig. 14.1 ). Therefore, a low CFR value may be a result of significant epicardial CAD, microvascular dysfunction, or both. For all of these reasons, invasively measured CFR has largely been abandoned as a method for interrogating intermediate coronary lesions. However, in patients with normal appearing epicardial coronary vessels, invasively measured CFR can be used to assess microvascular function. However, because of the previously mentioned limitations and the availability of other methods for assessing the microvasculature independently of the epicardial system (for example, the index of microcirculatory resistance, which will be discussed later), invasively measured CFR is not performed routinely on a clinical basis.
Fractional Flow Reserve
Because of the issues surrounding CFR mentioned earlier, in the early 1990s Pijls, De Bruyne, et al. introduced fractional flow reserve (FFR) as a method for assessing the functional significance of epicardial CAD. FFR is defined as the maximum myocardial blood flow in the presence of an epicardial stenosis compared with the maximum flow in the hypothetical absence of the stenosis. During maximal hyperemia microvascular resistance is minimized and assumed to be similar in the presence and absence of an epicardial stenosis. Therefore, flow becomes proportional to pressure, and the definition for FFR can be stated as the distal pressure in the presence of a stenosis compared with the distal pressure in the theoretical absence of the stenosis. In a normal epicardial vessel, distal coronary pressure is similar to proximal coronary pressure. Therefore, in a diseased epicardial vessel, what the distal coronary pressure would be in the absence of the disease can be approximated by measuring the proximal coronary pressure. This concept allows for measurement of FFR invasively by using a coronary pressure wire to measure mean distal pressure during maximal hyperemia and dividing that by the mean proximal coronary or aortic pressure measured simultaneously with the guiding catheter ( Fig. 14.2 ).
FFR has a number of unique attributes that makes it more attractive than CFR for assessing epicardial CAD ( Box 14.1 ). First, it has a normal value of 1.0 in every patient and every vessel. Second, it has a well-defined cut-off value of 0.75, with a “gray” zone extending to 0.80. If the FFR value is above 0.80, then it can be assumed that the epicardial vessel being interrogated is not responsible for significant ischemia. FFR values below 0.75 indicate that epicardial vessel disease is responsible for ischemia, whereas values in the gray zone require clinical judgment. It is important to remember that FFR is not a dichotomous variable, but a continuous one. In the same vessel, the lower the FFR value, the greater the degree of myocardial ischemia present, and the greater the benefit of revascularization compared with medical therapy.
- 1.
Normal value of 1.0 in every patient and every vessel
- 2.
Well-defined ischemic cut-off value
- 3.
Independent of hemodynamic perturbations
- 4.
Extremely reproducible
- 5.
Relatively easy to measure
- 6.
Specific for the epicardial vessel
- 7.
Independent of the microvasculature
A third attribute of FFR is that, because it is measured during maximal hyperemia, it is independent of changes in resting flow and other hemodynamic perturbations. For this reason, FFR has excellent reproducibility. Fourth, FFR is relatively easy to measure, at least in comparison to Doppler-derived CFR. Finally, FFR is a specific measure of the contribution of the epicardial CAD to myocardial ischemia. It is independent of microvascular dysfunction. This is an important advantage during invasive assessment because it provides information regarding the expected improvement in myocardial flow should a stent be placed across an epicardial stenosis. For example, in a vessel that subtends previously infarcted myocardium, the maximum flow down the vessel will be less than expected, leading to a lower gradient and a higher FFR across a given upstream epicardial stenosis. However, this does not mean that the FFR is inaccurate, it simply means that the epicardial stenosis does not have a significant effect on myocardial flow and is not responsible for myocardial ischemia.
Data Supporting the Clinical Utility of FFR Measurement
FFR was first validated in a landmark study by Pijls, De Bruyne, et al. in which FFR was compared with three different noninvasive stress tests for the assessment of ischemia in patients with intermediate coronary narrowings. If any one of the stress tests was positive for ischemia, the patient was defined as having ischemia. By using composite information from all three stress tests, the authors were able to increase the accuracy of the noninvasive diagnosis of ischemia. Using a cut-off point of 0.75, they found that 100% of the 21 patients with an FFR below 0.75 had ischemia and 88% of the 24 patients with an FFR of 0.75 or greater did not have ischemia. Importantly, revascularization was not performed in these 24 patients and at an average of 14-month follow-up there were no cardiac events in this group. The overall accuracy of FFR for identifying ischemia-producing lesions in patients with single-vessel intermediate disease was 93%.
The clinical utility of FFR has been documented in three large randomized trials and multiple registries and observational studies. The first important randomized study validating FFR was the deferral of percutaneous coronary intervention (DEFER) trial. In this study, 325 patients with chest pain and moderate coronary lesions who were scheduled for percutaneous coronary intervention (PCI) had FFR measured. If the FFR was less than 0.75, the patient was thought to have a functionally significant coronary stenosis and underwent PCI. If the FFR was greater than or equal to 0.75, the patient was randomized to performance of PCI anyway or to deferral of PCI with medical treatment. At 2-year follow-up, the major adverse cardiac event rate was 11% in the 91 patients randomized to deferral of PCI compared with 17% ( p = 0.27) in the 90 patients randomized to performance of PCI. This initial report was important because it was the first large-scale study to demonstrate the safety of deferring PCI for lesions that appeared to be associated with significant epicardial obstruction but that were not functionally significant based on FFR.
Subsequently, this same group of patients was followed up to 5 years, at which point in time the cardiac death and myocardial infarction rate was 3.3% in the deferral group versus 7.9% in the performance group ( p = 0.21), further documenting safety of medical treatment of coronary lesions that are not hemodynamically significant. Most recently, the 15-year follow-up of this cohort was published. At 15 years, the mortality rate was no different between the two groups, but the myocardial infarction rate was significantly lower in the deferral group compared with the performance group (2.2% vs 10%, p = 0.03). During follow-up this benefit came without any significant difference in revascularization between the two groups (44% vs 34%, respectively, p = 0.25). This report reinforced the safety of medically managing CAD that is not functionally significant based on FFR ( Table 14.1 ).
| Adverse Event | Deferral ( n = 91) | PCI ( n = 90) |
|---|---|---|
| 2-Year Follow-Up | ||
| Death (%) | 4.4 | 2.2 |
| MI (%) | 0 | 3.3 |
| Revascularization (%) | 5.6 | 7.8 |
| 5-Year Follow-Up | ||
| Death (%) | 6.6 | 5.7 |
| MI (%) | 0 | 5.6 |
| Revascularization (%) | 10.0 | 15.9 |
| 15-Year Follow-Up | ||
| Death (%) | 33.0 | 31.1 |
| MI (%) | 2.2 | 10.0 a |
| Revascularization (%) | 36.3 | 27.8 |
The next important multicenter, randomized study, the Fractional Flow Reserve Versus Angiography for Multivessel Evaluation (FAME) trial, established the role of FFR in guiding PCI in patients with multivessel CAD. FAME included 1005 stable and unstable patients (not including those with acute ST elevation myocardial infarction [STEMI]) who had 50% or greater stenoses in at least two of the major epicardial arteries and were thought to require PCI based on the angiographic appearance and the clinical scenario. After the identification of which lesions required PCI, the patients were randomized to either angiography-guided PCI, in which case PCI was performed in the usual fashion, or to FFR-guided PCI, in which case first FFR was measured across each lesion and then only if the FFR was less than or equal to 0.80 was PCI performed. The primary endpoint was the 1-year rate of death, myocardial infarction, or repeat revascularization.
Significantly fewer drug-eluting stents were placed in the group randomized to FFR-guidance than in those receiving usual care (1.9 vs 2.7 stents per patient, p < 0.001) despite a similar number of lesions identified in both groups. Importantly, the procedure time was similar between both groups; although measurement of FFR adds some time, avoiding unnecessary stents saves time. Most importantly, outcomes were improved in the FFR-guided group at 1 year with a significantly lower rate of the primary endpoint (13% vs 18%, p = 0.02). In addition, the rate of death or myocardial infarction was also significantly reduced in the FFR-guided patients (7% vs 11%, p = 0.04). Both 2-year and 5-year follow-up confirmed the durability of these results. Additionally, a cost-effectiveness study found that FFR-guided PCI was unique in that it not only improved outcomes but also saved resources ( Fig. 14.3 ).
The FAME trial was important because it highlighted the benefit of a new concept, functional revascularization, which involves performing PCI on only those lesions responsible for myocardial ischemia, as directed by an abnormal FFR, and treating medically those lesions that are not functionally significant, despite their angiographic appearance. In this manner the benefits of PCI are maximized while its risks are minimized.
An angiographic substudy of the FAME trial further emphasized the discordance between the coronary angiogram and the functional significance of lesions based on FFR. Of the lesions that were graded between 50% and 70% by the operator, 35% had an ischemic FFR, whereas for those between 70% and 90%, 20% were not significant based on FFR. This discrepancy is especially relevant in patients with three-vessel CAD for whom coronary artery bypass graft surgery is being considered. Guidelines suggest calculating the SYNTAX score and favor bypass surgery if the SYNTAX score is in the intermediate or highest tertile. However, the SYNTAX score is inherently limited by the fact that it is based on the angiographic appearance of the CAD and does not take into account the functional significance. To address this limitation, a so-called Functional SYNTAX Score has been proposed and evaluated in the cohort assigned to FFR-guided PCI in the FAME trial.
The Functional SYNTAX Score takes into account only those lesions that are significant based on FFR. It was shown to be a better discriminator of the risk for death and myocardial infarction as compared with the SYNTAX score ( Fig. 14.4 ). This concept is being tested prospectively in the FAME 3 trial, comparing FFR-guided PCI with current generation drug-eluting stents to coronary artery bypass graft surgery in a multicenter, randomized trial.
