13 Nonangiographic Coronary Lesion Assessment FFR, IVUS, OCT, NIRS

Morton J. Kern, John M. Hodgson, Arnold H. Seto

The rationale for using nonangiographic lesion assessment tools arises from two principles: (1) that revascularization (via percutaneous coronary intervention [PCI] or coronary artery bypass graft [CABG]) is justified best by the presence of ischemia, which depends on the hemodynamic significance of a lesion; and (2) that the coronary angiogram frequently fails to establish the hemodynamic significance of coronary stenoses with accuracy, particularly the intermediately narrowed (between 30% and 80% diameter stenosis) lesions. This limitation of angiography has been documented repeatedly by poor correlations to stress testing and is attributable to the anatomical complexity of the atherosclerotic lumen (Fig. 13-1).

Figure 13-1 A, Frame from cineangiogram of left coronary artery (LCA) with an intermediate left anterior descending (LAD) lesion. Arrow indicate LAD narrowing. B, Diagram of regions of stenosis resistance causing poststenotic pressure loss. (1) Entrance angle, (2) Length of disease, (3) Length of stenosis, (4) Minimal lumen diameter, (5) Minimal lumen area, (6) Eccentricity of lesion, (7) Area of reference vessel segment.

Coronary angiography can only produce a two-dimensional silhouette image of the three-dimensional vascular lumen. Angiographic accuracy is further limited by the inability to identify diffusely “diseased” and “normal” vessel segments. In addition, unlike intravascular ultrasound (IVUS), angiography does not provide vascular wall detail sufficient to characterize plaque size, length, and eccentricity. The eccentric lumen produces conflicting degrees of angiographic narrowing when viewed from different angulations, causing uncertainty related to lumen size and its impact on coronary blood flow. Moreover, there are at least six morphologic features that determine resistance to flow, most of which can be measured from the angiogram or even IVUS (Figs. 13-1, 13-2). Additional artifacts, including contrast streaming, branch overlap, vessel foreshortening, calcifications, and ostial origins, further contribute to uncertain angiographic lesion interpretation.

Figure 13-2 Diagram of angiographic projections demonstrating markedly different diameter narrowings illustrating the greatest limitation of angiography for eccentric lesions (left). Orthogonal projections of different orifice configurations that complicate determination of physiologic impact of narrowing (right).

The uncertainty of angiographic lesion assessment is a significant clinical problem. When evidence of ischemia is lacking, before stenting all intermediate lesions indiscriminately, the functional significance of a stenosis by fractional flow reserve should be identified as the first step in PCI. After the lesion is shown to be flow limiting, the anatomical and morphologic features of the stenosis and reference vessel segment can be assessed by IVUS. More detail on structure and composition can be obtained by optical coherence tomography (OCT) and, in the future, a determination of the plaque character (e.g., lipid pool content with near-infrared spectroscopy [NIRS]) may assist in appropriate stenting in regions beyond the most stenotic segment.

Thus, the three most common technologies for nonangiographic coronary lesion assessment tools available at this time are the (1) coronary pressure wire, (2) IVUS, and (3) OCT. The Doppler flow wire as a research tool to study the microcirculation and new imaging modalities such as NIRS imaging will be addressed briefly.

Coronary Pressure and Fractional Flow Reserve

Pijls and De Bruyne developed and validated an index for determining the physiologic impact of coronary stenoses, called the fractional flow reserve (FFR). FFR is measured as the ratio of mean distal coronary pressure divided by the mean proximal aortic pressure during maximal hyperemia. The coronary pressure beyond the stenosis is measured with a 0.014-inch guidewire with a high-fidelity pressure transducer mounted 3 cm from the tip of the wire, at the junction of the radiopaque and radiolucent segments. A full discussion of the FFR method and results can be found elsewhere (see Suggested Readings).

Concept of FFR

FFR is defined as the ratio of maximal hyperemic flow across an epicardial coronary stenosis compared with maximal hyperemic flow in the same artery without the stenosis. FFR is expressed as the percentage of normal maximal flow through the stenotic artery. FFR can be separately computed for the myocardium (FFR_m), the epicardial coronary artery (FFR_c), and the collaterals, (FFR_collat), based on translesional pressure measured during maximal hyperemia and in some cases coronary occlusion wedge pressure. Figures 13-3 and 13-4 illustrate the concept and data used to derive FFR. Table 13-1 lists the calculations for FFR and Table 13-2 lists the thresholds for clinical applications of FFR.

Figure 13-3 A, Diagrams of the theory of fractional flow reserve (FFR). FFR is the ratio of maximal myocardial perfusion in the stenotic territory divided by maximal hyperemic flow in that same region in the hypothetical case the lesion was not present (dotted artery behind solid lines). FFR represents that fraction of hyperemic flow that persists despite the presence of the stenosis. This ratio of two flows is calculated solely from the ratio of mean coronary pressure (Pd) divided by mean aortic pressure (Pa) provided both pressures are recorded under conditions of maximal hyperemia. Pa is the same along the length of the normal vessel. FFR is defined as myocardial flow (Qs) across stenosis/myocardial flow (Qn) without stenosis. To derive FFR, assume resistance = P/Q, then flow, Q = P/R, and that Qs/Qn = (Pd/Rs)/(Pa/Rn), where Rs, Rn is resistance in stenotic and normal bed which are identical at maximal hyperemia. If Rs = Rn, then Qs/Qn = Pd/Pa, which is FFR = Qs/Qn = Pd/Pa. B, Pressure signals used to calculate FFR. Pa and Pd are recorded at rest and then during hyperemia induced by adenosine, in this case intracoronary. The nadir of distal pressure is used for the FFR calculation. Shown under the yellow band is the flow velocity signal, which illustrates that the peak flow corresponds to the nadir of distal pressure. C, Screen from FFR monitor showing colored signals of Pa (red) and Pd (yellow) for FFR of 0.86. D, Hemodynamic display of FFR across severe stenosis with FFR of 0.40. Note very low distal pressure.

Figure 13-4 A (top pannel) pressure and flow velocity signals used to measure FFR. CVR = coronary vasodilatory reserve. B (lower pannel) Translesional pressure measurements before (top) and after (bottom) LAD stent placement. Resting gradients do not always indicate severity of lesion. Prestent FFR = 0.55 despite small resting gradient. Poststent FFR = 0.98, a normal value.

(Courtesy of Dr. Bernard DeBruyne.)

^{Table 13-1} Calculations of Fraction Flow Reserve From Pressure Measurements Taken During Maximal Arterial Vasodilation

Myocardial fraction flow reserve (FFR_myo):
FFR_myo= 1 – ΔP/P_a – P_v
= P_c – P_v/P_a – P_v
= P_c/P_a
Coronary fractional flow reserve (FFR_cor): FFR_cor = 1 – ΔP(P_a – P_w)
Collateral fractional flow reserve (FFR_coll): FFR_coll = FFR_myo – FFR_cor

Note: All measurements are made during hyperemia except P_w.

P_a, mean aortic pressure; P_c, distal coronary pressure; ΔP, mean translesional pressure gradient; P_v, mean right atrial pressure; P_w, mean coronary wedge pressure or distal coronary pressure during balloon inflation.

From Pijls NHJ, van Som AM, Kirkeeide RL, 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 1993;87:1354–1367.

Table 13-2 Physiologic Criteria Associated With Clinical Applications

FFR differs from absolute coronary flow reserve (CFR, maximal flow/basal flow) since it does not depend on basal flow levels but is computed only at maximal flow (hyperemia). FFR has several advantages over CFR:

1. It has an absolute normal value of 1.0 for every artery, every patient.

2. It is not affected by changing hemodynamics or status of the microcirculation.

3. It is specific for epicardial coronary stenoses.

Technique of FFR

FFR can be easily measured using a 5F or 6F guide catheter and either of two available pressure wire systems (St. Jude Medical, Minneapolis, MN or Volcano Therapeutics, Rancho Cordova, CA). After diagnostic angiography with a catheter seated in the coronary ostium, the steps to measure FFR are as follows:

1. The pressure wire is connected to the system’s pressure analyzer and calibrated and zeroed outside the body.

2. Anticoagulation intravenous (IV) heparin (usually 40 U/kg) and intracoronary (IC) nitroglycerin (100–200 μg bolus) are administered.

3. The wire is advanced through the guide to the coronary artery. The pressure wire signal and the guide pressure are matched (i.e., equalized, also called normalized) before crossing the stenosis. By early convention, the guidewire transducer was positioned at the end of the guide catheter. In fact, it does not matter exactly where the wire is in relation to the guide catheter or coronary ostium except that both should be in the aortic sinus when equalizing the signals.

4. The wire is then advanced across the stenosis about 2 cm distal to the coronary lesion.

5. Maximal hyperemia is induced with IV adenosine (140 μg/kg/min) or IC bolus adenosine (20–30 μg for the right coronary artery, 60–100 μg for the left coronary artery). Alternative hyperemic agents are rarely used but include IC papaverine (10 mg), nitroprusside (50–100 μg), or ATP (50–100 μg). FFR is measured at 2 min for IV adenosine and at 15 to 20 seconds after IC adenosine.

6. The ratio of the mean distal pressure to mean proximal pressure during maximal hyperemia is calculated as the FFR. An FFR <0.80 has a strong ischemic correlation and is an indication to proceed with PCI. If a PCI is deemed necessary, it can be performed using the pressure wire as the angioplasty guidewire. After the procedure, FFR can be remeasured to assess the adequacy of the intervention.

7. Finally, at the end of the procedure (either the assessment or the PCI), the pressure wire should be pulled back into the guide to confirm equal pressure readings, or the lack of pressure wire drift. Table 13-2 shows the criteria for FFR and associated clinical situations.

Coronary Hyperemia for FFR

IV adenosine is weight based, operator independent, and the preferred method of inducing hyperemia. By providing a prolonged hyperemic stimulus, IV adenosine allows for a slow pullback of the pressure wire, useful for identifying the exact location of the pressure dropoff or the presence of diffuse disease. It is also required for assessment of aorto-ostial narrowings to permit maximal coronary flow without guide catheter obstruction (Box 13-1, Table 13-3A, Box 13-2, Table 13-3B).

Box 13-1 Intracoronary Adenosine

(Modified from Volcano Therapeutics.)

^{Table 13-3A} Intracoronary Adenosine

Effects:
• Peak effect:	≤10 seconds after administration.
• Duration of plateau:	<20 seconds.
Side Effects:
• AV block.	Short and rapidly transient. Usually after injection in the RCA.
Comments:
Hyperemia generally
• Just before inducing hyperemia, give nitrates, if appropriate for the patient, as per regular coronary intervention.	Avoids performing measurements influenced by spasm.
I.C. bolus injection generally
• Guide catheters with side holes should NOT be used.	Unknown amounts of the drug may spill into the aorta.
• Guide catheters that are too large (tight) for the ostium should NOT be used.	Pressure damping may occur. This can be recognized by a ventricularized aortic pressure curve.
I.C. adenosine specifically
• NO pullback curve possible.	No steady-state hyperemia.
	Overestimation of FFR (underestimation of stenosis severity) may occur if only mean pressure is recorded.
• Interruption of aortic pressure (Pa) should be as short as possible.	If too long, hyperemia will be over before aortic pressure can be measured again.
• Successive measurements using IC adenosine may be performed.	The effects of adenosine wear off quickly. Wait between measurements just long enough for the previous dose to cease to have effect.
• Ensure that maximum effects of the hyperemic stimulus have been achieved.	In case of suboptimal hyperemia, overestimation of FFR (underestimation of stenosis severity) may occur. If FFR 0.75-0.85, check for pitfalls and repeat measurement, increase dose or switch to I.V. adenosine or another drug. In 10-15% of patients only submaximum hyperemia can be achieved using I.C. adenosine. FFR may be underestimated due to e.g., caffeine and theophylline (adenosine are inhibited by adenosine receptor antagonists, such as methylxanthines).

(Modified from Volcano Therapeutics.)

Box 13-2 Intravenous Adenosine

(Modified from Volcano Therapeutics.)

^{Table 13-3B} Intravenous Adenosine

Effects:
• Peak effect:	≤2 minutes after administration in central vein.
• Duration of effect:	Effect disappears within 2 minutes after infusion stopped.
• The effects of Adenosine Triphosphate (ATP) are condsidered equivalent to adenosine.
Side Effects:
• AV block.	Rarely.
• Do NOT use in patients with bronchoconstrictive or bronchospastic lung disease (e.g., asthma).	Risk for bronchospasm.
• Decrease in blood pressure and increase in heart rate by 10-20%	A slight increase in blood pressure often precedes the decrease.
• Unpleasant angina-like or burning sensation in chest or throat during infusion.	Disappears rapidly after ending the infusion. Harmless and does not indicate ischemia. Inform and reassure the patient in advance.
Comments:
Hyperemia generally:
• Just before inducing hyperemia, give nitrates, if appropriate for the patient, as per regular coronary intervention.	Avoids performing measurements influenced by spasm.
I.V. infusion generally
• Use volume-controlled infusion pump with sufficient capacity.	Inadequate infusion may result in suboptimal hyperemia.
• Infuse in femoral or large antecubital vein. If brachial vein is used for infusion, avoid kinking of brachial vein caused by bending of the patient’s forearm. Patient’s arm should be extended.	Improper infusion may result in fluctuations in pressure.
I.V. adenosine specifically
• Pullback curve possible.	Steady-state hyperemia.
• Instruct patient to breath normally. The patient should avoid Valsalva-like maneuvers.	Decrease of venous return = pressure signal fluctuations—overestimation of FFR.
• Ensure that maximum effects of the hyperemic stimulus have been achieved.	In case of suboptimal hyperemia, overestimation of FFR (underestimation of stenosis severity) may occur. If FFR 0.75-0.85, check for pitfalls and repeat measurement, increase dose, or switch to another drug. Only submaximum hyperemia can be achieved in approximately 8% of patients using IV adenosine. FFR may be underestimated due to e.g., caffeine and theophylline (adenosine is inhibited by adenosine receptor antagonists, such as methylxanthines).

(Modified from Volcano Therapeutics.)

While IC adenosine is equivalent to IV infusion for determination of FFR in a large majority of patients, in a small percentage of cases, coronary hyperemia may be suboptimal with IC adenosine. Jeremias et al. compared IC (15–20 μg in the right, and 18–24 μg in the left coronary artery) with IV adenosine (140 μg/kg/min) in 52 patients with 60 lesions. There was a strong and linear relationship between IC and IV adenosine (r = 0.978 and P < 0.001). The mean measurement difference for FFR was − 0.004 ± 0.03. In 8.3% of stenoses, FFR with IC adenosine differed by 0.05 or more compared with IV adenosine, suggesting an inadequate hyperemic response.

Pitfalls of FFR

As a cautionary note, catheters with side holes should not be used to measure FFR, as proximal pressure gradients may occur, complicating distal gradient evaluations. Larger guide catheters can partially occlude the coronary ostium as hyperemia is induced, impairing maximal flow. Removing the guide catheter from the coronary ostium after giving the hyperemic agent will avoid this pitfall.

The errors in the performance of accurate FFR involve hemodynamic artifacts and failure to induce maximal hyperemia. Tables 13-4, 13-5, and 13-6 list these problems. Figures 13-5, 13-6, 13-7, and 13-8 show artifacts that may produce false FFR readings.

^{Table 13-4} Reasons for Nonischemic Fractional Flow Reserve (FFR) Despite an Apparently Severe Stenosis

Physiologic Explanations
Stenosis hemodynamically nonsignificant despite angiographic appearance Small perfusion territory, old myocardial infarction, little viable tissue, small vessel Abundant collaterals Severe microvascular disease (rarely affecting FFR)
Interpretable Explanations
Other culprit lesion, diffuse disease not focal stenosis Chest pain of noncardiac origin
Technical Explanations
Insufficient hyperemia Guiding catheter related pitfall (deep engagement, small ostium, side holes) Electrical drift
Actual False Negative FFR
Acute phase of ST-segment elevation myocardial infarction Severe left ventricular hypertrophy Exercise-induced spasm

From Koolen JJ, Pijls NHJ. Coronary pressure never lies. Catheter Cardiovasc Interv 2008;72:248–256.

^{Table 13-5} Factors Involved in Fractional Flow Reserve Accuracy

Hemodynamic Artifacts or Errors
1. Signal drift. This is rare and can be determined by careful observation of pressure waveform and presence of dicrotic notch. If suspicious, check with re-matching of signals on pullback to aortic location. 2. Incorrect height of pressure transducer. 3. Loss of pressure due to guidewire introducer. Remove and tighten Touhey-Borst valve. 4. Damping of pressure by guiding catheter. Observe aortic pressure wave. Use IV adenosine and keep catheter in aorta, not coronary ostium. 5. Guiding catheters with side holes produce pseudostenosis across the catheter into the coronary ostium. Use IV adenosine and keep catheter in aorta, not coronary ostium. 6. Pressure damping with 4F or 5F catheters if not flushed with saline. Contrast media viscosity will produce unreliable aortic pressure wave.
Failure to Induce Hyperemia
A. Adenosine, intracoronary bolus administration 1. Submaximum stimulus in some patients. Very rare; if suspected select alternative agent (e.g., nitroprusside, papaverine, adenosine triphosphate [ATP]). 2. Failure to capture pressure change at peak hyperemia. Maximum gradient underestimated when calculated from mean signal, unless it is taken on beat-to-beat basis. 3. No pullback curve possible with bolus administration. 4. Guiding catheter failure to seat and delivery drug. 5. Guide catheter flow obstruction. 6. Incorrect dose mix or dilution. B. Adenosine, IV adenosine 1. Check infusion, pump system, and lines. 2. Infuse through central vein. 3. Avoid Valsalva maneuver during infusion. 4. Decrease of blood pressure by 10%–15%. 5. Burning or angina-like chest pain during infusion. This is a harmless effect and does not indicate ischemia. IV adenosine not to be used in patients with severe obstructive lung disease (bronchospasm). 6. If peripheral vein is used, avoid kinking of arm/elbow. 7. Avoid Valsalva maneuvers.

^{Table 13-6} Most Common Reasons for False Negative and False Positive Fractional Flow Reserve (FFR)

False Negative

Pressure Damping

No hyperemia

Wrong drug, not mixed not delivered (IV) or side holes

False Positive

Small artery, small territory

pressure signal drift

Cautionary considerations: pathophysiologic conditions theoretically limiting FFR

Left Ventricular Hypertrophy

Exercise-Induced Vasoconstriction

Microvascular Disease and Myocardial Infarction

Figure 13-5 Example of pressure signal damping (left) and after withdrawal of guide catheter (right).

Figure 13-6 Example of pressure signal drift as a cause for false positive fractional flow reserve (FFR). Note preservation of dicrotic notch on distal pressure indicating normal transmission of pressure.

Figure 13-7 Example of distal pressure across severe stenosis. Note distal pressure wave configuration with wide pulse pressure and loss of dicrotic notch indicative of severe stenosis.

Figure 13-8 Illustration of bed size on fractional flow reserve (FFR). Because flow is related to myocardial bed size, a moderate lesion (top, 50%) serving a large territory could have low FFR (0.75) while a more severe appearing lesion (bottom, 85%) serving a small bed (such as after infarction) could have high FFR (0.83).

Use of FFR for Specific Angiographic Subsets

Intermediate Coronary Lesion

FFR assists the operator in deciding to treat or not treat coronary lesions based on ischemia. Some operators are concerned that not stenting intermediate but hemodynamically insignificant lesions will result in harm to the patient later. This concern is unfounded based on the 5-year outcomes of the DEFER study. The DEFER study randomized 325 patients scheduled for PCI into three groups: a deferral group (n = 91) in whom an FFR was ≥ 0.75 and medical therapy was continued; or despite an FFR > 0.75, the PCI performance group (n = 90), in which the lesions were treated with stents. The third group was the reference group (n = 144) who had an FFR < 0.75 and stents were placed as planned. For the deferred and performed groups, the event-free survival was the same (80% and 73% respectively, P = 0.52), and both were significantly better than in the reference group (63%, P = 0.03). The composite rate of cardiac death and acute myocardial infarction (MI) in the deferred, performed, and reference groups was 3%, 8%, and 16% respectively (P = 0.21 for deferred vs. performed and P = 0.003 for reference vs. both of the deferred and performed groups) (Fig. 13-9). The percentage of patients free from chest pain on follow-up was not different between the deferred and performed groups. The 5-year risk of cardiac death or MI in patients with a normal FFR is <1% per year and is not decreased by stenting. Treating patients with intermediate lesions assisted by FFR is associated with a low event rate, comparable to event rates in patients with normal noninvasive testing. Figure 13-10 is an example of FFR for intermediate lesion assessment. Similar outcomes for deferment of lesions with FFR > 0.80 were also reported in patients in the FAME study described in the next section.

Figure 13-9 The DEFER study results. Kaplan-Meier survival curves for freedom from adverse cardiac events during 5 years of follow-up for the defer group (blue line), treatment group (red line), and reference group (black line).

(From Pijls NHJ et al. J Am Coll Cardiol 2007;49:2105–2111.)

Figure 13-10 Angiograms of patient with intermediate LAD and severe OM1 branch lesion. A, LAO view of LAD (left), LAO caudal view of LCA (middle), and occluded RCA of 2 years ago (right). B, RAO cranial shows intermediate LAD (left), severe OM1 branch stenoses (right). FFR of LAD is 0.86, 0.87. C, Pre-PCI of OM1 (left) and post-PCI (right). Patient became pain free and did well. FFR, fractional flow reserve; LAD, left anterior descending; LAO, left anterior oblique; LCA, left coronary artery; OM1, first obtuse marginal; PCI, percutaneous coronary intervention; RAO, right anterior oblique; RCA, right coronary artery.

Multivessel Disease PCI

The FAME (FFR vs. Angiography for Multivessel Evaluation) trial by Tonino et al. compared a physiologically guided PCI approach (FFR-PCI) to a conventional angiographic guided PCI (Angio-PCI) in patients with multivessel coronary artery disease (CAD). A total of 1005 patients with multivessel CAD undergoing PCI with drug-eluting stents were enrolled. Operators identified all lesions by visual angiographic appearance (>50% diameter stenosis) to be treated in advance of randomization to a stenting strategy. For the FFR-PCI group (n = 496), all lesions had FFR measurements and only those with FFR < 0.80 were stented. For the Angio-PCI group (n = 509), all lesions identified were stented. Clinical characteristics and angiographic findings were similar in both groups with average SYNTAX scores of 14.5 (indicating low-intermediate risk patients).

Compared with the Angio-PCI group, the FFR-PCI group used fewer stents per patient (1.9 ± 1.3 vs. 2.7 ± 1.2, P < 0.001) and less contrast (272 mL vs. 302 mL, P < 0.001), had a lower procedure cost (US$ 5,332 vs. US$ 6,007, P < 0.001), and had a shorter hospital stay (3.4 vs. 3.7 days, P = 0.05). More importantly, the 2-year rates of mortality or MI were 13% in the Angio-PCI group compared with 8% in the FFR-PCI group (P = 0.02). Composite rates of death, nonfatal MI, or revascularization were 22% and 18%, respectively (P = 0.08). For lesions deferred on the basis of FFR > 0.80, the rate of MI was only 0.2% and the rate of revascularization was 3.2 % after 2 years (Fig. 13-11).

Figure 13-11 The FAME study results. Kaplan-Meier survival curves according to study group for 2-year outcomes. CABG, coronary artery bypass graft; FFR, fractional flow reserve; MACE, major adverse cardiac event; MI, myocardial infarction; PCI, percutaneous coronary intervention.

(From Tonino PAL et al. N Engl J Med 2009;360:213–224.)

FAME demonstrated that PCI guided by FFR in patients with multivessel CAD significantly reduces mortality and MI at 2 years when compared with standard angiography-guided PCI. A related cost-effectiveness evaluation showed that FFR-guided PCI not only improved outcomes, but did so at a significantly lower cost.

Significance of Abnormal FFR after PCI

FFR after bare-metal stenting predicts adverse cardiac events at follow-up. Pijls et al. examined 750 patients with angiographically satisfactory PCI using postprocedural FFR. At 6 months, 76 patients (10%) suffered an adverse event. FFR immediately after stenting was the most significant independent variable related to all types of events. In 36% of patients, FFR normalized (>0.95) and patients had an event rate of only 5%. In 32% of patients with postprocedure FFR between 0.90 and 0.95, the event rate was 6%. In the remaining 32% with FFR <0.90, the event rate was 20%, and was 30% among those patients with FFR <0.80. These data suggest that both edge stent abnormalities (e.g., occult dissection) and diffuse disease are associated with a worse long-term outcome after bare-metal stenting. Outcome correlations with FFR after drug-eluting stenting have not yet been reported.

Left Main Stenosis

Accurate assessment of the hemodynamic significance of left main coronary lesions is of critical importance when patients face possible CABG surgery. Because of the inherent limitations discussed earlier, angiography alone may not be reliable in intermediate left main stenoses, and FFR is useful for decision making.

Numerous studies of FFR support its use in equivocal left main CAD (Table 13-7). Most recently, Hamilos et al., in a large multicenter prospective trial, examined FFR and 5-year outcomes in 213 patients with an angiographically equivocal left main coronary artery stenosis. When FFR was >0.80, patients were treated medically or another stenosis was treated by coronary angioplasty (nonsurgical group; n = 138). When FFR was <0.80, CABG surgery was performed (surgical group; n = 75). The 5-year survival estimates were 90% in the nonsurgical (FFR > 0.80) group and 85% in the surgical (FFR < 0.80) group (P = 0.48). The 5-year event-free survival estimates were 74% and 82% in the two groups, respectively (P = 0.50) (Fig. 13-12). Of note, only 23% of patients with a diameter stenosis >50% had a hemodynamically significant left main stenosis as measured by FFR. Table 13-7 lists studies with FFR for left main stenosis.

Table 13-7 Fractional Flow Reserve (FFR) to Assess Intermediate Left Main Coronary Stenoses