To obtain these measures in the catheterization laboratory a pressure wire is placed across the stenosis, an adequate hyperemic stimulus administered, and one recording of the proximal and distal pressure is obtained to calculate the FFR (Fig. 3). Two phasic and two mean signals are obtained. One is recorded by the guiding catheter (aortic pressure), and the other by the pressure sensor, across the stenosis.

In cross-sectional lesions, clear pressure decreases are seen. Information about stenosis severity and exact lesion location are provided by the registrations. When overprojection or other angiographic problems cause difficulty in interpretation, hemodynamic information can determine the location of the problem. Essential for this concept is the presence of maximal hyperemia, required to unmask the true severity of a stenosis. When flow is increased, the distal pressure is decreased (Fig. 4), and FFR can be calculated by taking the mean distal pressure divided by the mean proximal pressure. Useful stimuli include intracoronary papaverine, intracoronary adenosine, intravenous adenosine, intracoronary ATP, and intravenous ATP. The presently used doses are higher than those recommended in the past, but if used carefully valuable measures of FFR can be obtained.

An example with intravenous adenosine showed that the small resting gradient increases to a large gradient of hyperemia and good steady state, allowing calculation of the FFR (Fig. 5). In diffuse disease or for multiple lesions along a coronary artery, this method of inducing hyperemia is convenient as it allows for starting in the distal coronary artery with the pressure sensor and making a slow pullback curve during continuous hyperemia, yielding good quality information about whether disease is present and its location.

Similar results are achieved with the difference hyperemic stimuli, regardless of route or site of delivery, based on Pd/Pa ratios. Only the contrast agent is an inadequate stimulus and should not be used as an alternative to adenosine.

The theoretical unequivocal normal value for FFR is 1.0. This value has been validated by Pijls and colleagues in their study of FFR in 33 coronary arteries in 8 normal persons. Their coronary angiograms, dobutamine echocardiograms and LV function were all normal. In a normal coronary artery, there is no decline of pressure from the proximal to distal part of the coronary artery, and the normal FFR range was found to be 0.98 + 0.2 (range 0.94-1.02).

FFR is independent of changes in blood pressure and heart rate. CFR is defined as maximum blood flow in the presence of stenosis divided by baseline blood flow. If blood pressure decreases, CFR will change to A prime divided by B prime. If heart rate increases, the baseline pressure-flow relationship level increases, and the CFR is A divided by B prime squared. CFR will change, although the stenosis does not change (Fig. 6, left panel). FFR is defined as maximum blood flow in the presence of a stenosis divided by normal maximal blood flow. If blood pressure decreases, A prime is divided by C prime, which is still the same ratio. Heart rate activity increases do not affect this ratio, and has been validated in patients with heart rates between 50 and 150 beats per minute (Fig. 6, right panel).

FFR values and ischemia

An FFR below 0.75 is associated with inducible ischemia if the patient is sufficiently stressed. For the most part, inducible ischemia is not possible with an FFR above 0.75, with a specificity of 100% and sensitivity of 90%. In 10% of patients, it may be possible to induce ischemia with an FFR above 0.75. A number of diseases contribute to this, including microvascular disease.

FFR can be useful to determine whether a coronary intervention will beneficial. For example, in a patient with an intermediate left anterior descending (LAD) stenosis, two equal signals were recorded at one location, showing no resting gradient. With hyperemia only a small gradient develops (FFR = 0.89). This lesion is not responsible for the angina.

Is it safe to defer percutaneous coronary angioplasty for such a lesion? Thallium testing has shown that the strongest prognostic predictor is the presence and extent of inducible ischemia. Thus, the past studies suggest that only hemodynamically significant lesions should be dilated. The DEFER study was conducted to investigate this question with invasive measurements. This prospective, randomized study was conducted in 325 patients with an equivocal stenosis in a large coronary artery. In the 40% of patients with an FFR less than 0.75, a PTCA was performed. The patients with an FFR above 0.75 were randomized to either receiving a PTCA or PTCA was deferred.

Interestingly, the survival curve of the reference group (patients with an FFR less than 0.75) showed that the one year event-free survival was 81%. About 20% of these patients have either an event or revascularization within the next year. The one-year recurrence rate after single vessel PTCA is 20%, which is acceptable as the patients receive much functional benefit from PTCA. This was also seen in the reference group (FFR <0.75) in the DEFER study, with only 10% being angina-free at the beginning of the study and more than 70% were angina-free after one year. In a normal population at the same age, the expected one-year cardiac event rate is 1.5%. In the patients with a deferred PTCA, the event free survival was 93%. This 7% event rate is not high in absolute terms, but it is still 5-fold higher than in a healthy population. In the PTCA group, the event rate was 11%. Thus, these patients do not derive benefit from the angioplasty. Importantly, between the three groups there was no significant difference in functional class after one year.

The DEFER results show that a PTCA should be performed in patients with hemodynamically significant lesions to provide functional benefit, but this will not necessarily improve prognosis. If a lesion is not hemodynamically significantly, patients do not derive benefit from PTCA and medical treatment can be used which is safer and results in a better outcome.

FFR can also be used to evaluate balloon angioplasty. In patients with an FFR greater than 0.9, the 2-year restenosis rate was 2.5 fold lower than in those patients with an FFR less than 0.9. An FFR less than 0.75 represents an insufficient PTCA result. An FFR of 0.75-0.90 represents a suboptimal PTCA result and is associated with a one-year restenosis rate of 30%. Stenting could be considered. An FFR above 0.90 is a stent-like result and is associated with a one-year restenosis rate of 10%. This correlates well with the data from the DEBATE and DESTINI trials.

Stenting

A study of FFR after coronary stenting by Hanekamp and colleagues showed that a high post-stent FFR (greater than 0.94) corresponds to a good IVUS result. However, it was possible to obtain such a good result in only 64% of the patients. A post-stent FFR greater than 0.90 is associated with a moderate restenosis rate and was achievable in 84% of patients in this study. In the case of disease elsewhere in the vessel, the post-stent FFR should be represented by the pullback curve across the stent itself, because a residual gradient can also originate more distal or proximal to the stent. In a diagnostic setting, 0.75 is the threshold for intervention. A post-intervention FFR of at least 0.90 should be achieved. If this can not achieved with PTCA, stenting should be performed.

More complex coronary artery disease in which FFR has proven useful was reviewed. In a patient referred for PTCA for a mid-LAD lesion, because of concern of stenosis in another region, a pressure wire was placed far down into the LAD and a pullback curve was made during maximal hyperemia. In the distal LAD there was a large gradient with an FFR of 0.5. During sustained hyperemia there was a gradient at the site of the mid-LAD lesion, but a much larger gradient at the site of the proximal LAD lesion. Thus, the patient was referred to mid-CAB surgery, which was successful. It is possible to calculate the degree of influence of each of the lesions.

FFR can be especially useful in diffuse disease. In a patient with angina and a positive exercise test referred for PTCA, Pijls and colleagues first placed a pressure wire in the distal circumflex and the distal LAD to make pullback curves. In the proximal area of the left main, there was no pressure gradient. A large gradient in the distal LAD was found, and on the pullback curve a gradual decline of pressure. The pressure measurements inform that PTCA is not possible and medical therapy is warranted.

The acute phase of myocardial infarction is very dynamic. Pressure measurements are not useful during the first 48 hours of infarction. Treatment should be guided by clinical symptoms and ECG. After the artery has stabilized, pressure measurements can be useful. This was demonstrated in a 55-year old woman 6 days post Q-wave anterior infarction. On cardiac catheterization, the anterior wall was akinetic. Two questions must be answered: the amount of viable myocardium and whether the residual lesion is significant. Pressure measurements showed a resting gradient of 10 mm, and a large increase is seen after infusing adenosine, indicating preserved vascular reactivity in the anterior wall and viability. The rather low FFR indicates that the residual lesion is still hemodynamically significant. This has been validated in 40 patients. An FFR above 0.75 has been shown to be associated with a negative MIBI Spect scan.

Fractional collateral blood flow is the portion of the blood flow that goes to the myocardium through the collaterals, compared to that going through the coronary artery. . Coronary measures can distinguish these flow patterns. Fractional collateral flow is maximum recruitable collateral blood flow expressed as a fraction of normal maximal flow: (Pw - Pv/Pa - Pv). If this ratio is above approximately 0.3, there is a 6-fold lower chance of a myocardial infarction occurring during a follow-up of 3-5 years.