Previous studies demonstrated that coronary revascularization, especially percutaneous coronary intervention (PCI), does not significantly decrease the incidence of cardiac death or myocardial infarction in patients with stable coronary artery disease. Many studies using myocardial perfusion imaging (MPI) showed that, for patients with moderate to severe ischemia, revascularization is the preferred therapy for survival benefit, whereas for patients with no to mild ischemia, medical therapy is the main choice, and revascularization is associated with increased mortality. There is some evidence that revascularization in patients with no or mild ischemia is likely to result in worsened ischemia, which is associated with increased mortality. Studies using fractional flow reserve (FFR) demonstrate that ischemia-guided PCI is superior to angiography-guided PCI, and the presence of ischemia is the key to decision-making for PCI. Complementary use of noninvasive MPI and invasive FFR would be important to compensate for each method’s limitations. Recent studies of appropriateness criteria showed that, although PCI in the acute setting and coronary bypass surgery are properly performed in most patients, PCI in the non-acute setting is often inappropriate, and stress testing to identify myocardial ischemia is performed in less than half of patients. Also, some studies suggested that revascularization in an inappropriate setting is not associated with improved prognosis. Taken together, the presence and the extent of myocardial ischemia is a key factor in the management of patients with stable coronary artery disease, and coronary revascularization in the absence of myocardial ischemia is associated with worsened prognosis.

Coronary artery disease is a leading cause of mortality and morbidity in developing and developed countries[1-5]. In approximately half of patients with newly diagnosed coronary artery disease, the first presentation is either acute myocardial infarction or sudden cardiac death[6,7].

The development of percutaneous coronary intervention (PCI) has enhanced the management of patients with acute coronary syndrome, and the prognosis of these patients has been considerably improved[8-15]. However, in patients with stable coronary artery disease, coronary revascularization decreases angina symptoms but does not significantly prevent cardiac death or myocardial infarction[16-21]. Recent studies suggest that the presence and extent of myocardial ischemia determine the prognosis of patients with stable coronary artery disease. Coronary revascularization is associated with improved prognosis in patients with moderate or severe ischemia, but is associated with worsened prognosis in patients with no or mild ischemia[22,23]. In this article, studies with myocardial perfusion imaging (MPI) and fractional flow reserve (FFR) on the effects of coronary revascularization on prognosis are reviewed.

Previous studies demonstrated that coronary revascularization does not significantly decrease the incidence of cardiac death and myocardial infarction in patients with stable coronary artery disease[16-21]. In particular, the Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) study had a tremendous impact on our management of patients with stable coronary artery disease[24]. COURAGE trial is a randomized trial involving 2287 patients who had objective evidence of myocardial ischemia and significant coronary artery disease. The investigators assigned 1149 patients to undergo PCI with optimal medical therapy (PCI group) and 1138 to receive optimal medical therapy (OMT group) alone. The 4.6-year cumulative primary outcome (death from any cause and nonfatal myocardial infarction) rates were 19.0% in the PCI group and 18.5% in the OMT group (HR for the PCI group: 1.05; 95%CI: 0.87-1.27; P = 0.62). There were no significant differences between the PCI group and the OMT group in the composite of death, myocardial infarction, and stroke (20.0% vs 19.5%, HR = 1.05; 95%CI: 0.87-1.27; P = 0.62); hospitalization for acute coronary syndrome (12.4% vs 11.8%, HR = 1.07; 95%CI: 0.84-1.37; P = 0.56); or myocardial infarction (13.2% vs 12.3%, HR = 1.13; 95%CI: 0.89-1.43; P = 0.33). They concluded that as an initial management strategy in patients with stable coronary artery disease, PCI did not reduce the risk of death, myocardial infarction, or other major cardiovascular events when added to OMT.

However, the COURAGE Trial Nuclear Substudy tells another story[25]. This study enrolled 314 patients who underwent MPI performed before treatment and 6 to 18 mo after randomization. At follow-up, the reduction in ischemic myocardium was greater with PCI than with OMT (-2.7% vs -0.5%; P < 0.0001). More PCI patients exhibited significant ischemia reduction (33% vs 19%; P = 0.0004), especially patients with moderate to severe pretreatment ischemia (78% vs 52%; P = 0.007). Patients with ischemia reduction had lower ischemia-unadjusted risk of death or myocardial infarction (P = 0.037; risk-adjusted P = 0.26), particularly if baseline ischemia was moderate to severe (P = 0.001; risk-adjusted P = 0.08). Death or myocardial infarction rates ranged from 0% to 39% for patients with no residual ischemia to ≥ 10% residual ischemia on follow-up MPI (P = 0.002; risk-adjusted P = 0.09). Thus this study showed that adding PCI to OMT resulted in a greater reduction in ischemia compared with OMT alone, although the effect of PCI on death or myocardial infarction was borderline significant probably because of the small number of patients.

MPI is the most commonly used test to assess the presence and the extent of myocardial ischemia. Many studies demonstrated that the presence and extent of myocardial ischemia was closely related to adverse cardiac events[26-36]. Hachamovitch et al[36] identified 5183 patients who underwent MPI and were followed up for the occurrence of cardiac death or myocardial infarction. Over a mean follow-up of 642 ± 226 d, 119 cardiac deaths and 158 myocardial infarctions occurred, giving an annual cardiac death rate of 3.0% and annual myocardial infarction rate of 2.3%. In patients with no [summed stress score (SSS) 0-3], mild (SSS 4-8), moderate (SSS 9-13), and severe (SSS > 13) ischemia, the annual cardiac death rate was 0.3%, 0.8%, 2.3%, and 2.9%, respectively. Similarly, in patients with no, mild, moderate, and severe ischemia, the annual myocardial infarction rate was 0.5%, 2.7%, 2.9%, and 4.2%, respectively. Thus increased myocardial ischemia is associated with more frequent cardiac events.

Many studies also showed that coronary revascularization has a beneficial effect in patients with moderate to severe ischemia[22,23,37]. Hachamovitch et al[22] studied 10627 patients without known coronary artery disease who underwent MPI and were followed up for 1.9 ± 0.6 years. Within 60 d after MPI, 671 patients underwent revascularization therapy and 9956 patients underwent medical therapy (MT). On the basis of the Cox proportional hazards model predicting cardiac death, patients undergoing MT demonstrated a survival advantage over patients undergoing revascularization in the setting of no or mild ischemia (% total myocardial ischemia less than 10%), whereas patients undergoing revascularization had an increasing survival benefit over patients undergoing MT when moderate ischemia (% total myocardial ischemia 11%-20%) to severe ischemia (% total myocardial ischemia more than 20%) was present. In 2011, the same authors expanded their sample to 12329 patients and studied the interaction between the extent of ischemia and myocardial scar after revascularization on patient survival[23]. In the absence of prior coronary artery disease, increasing amounts of ischemia were associated with lower HRs with early revascularization. In the setting of little or no ischemia, early revascularization was associated with an approximately 50% greater risk than MT, whereas, with increasing ischemia, a progressive improvement in risk with early revascularization compared with MT was found. In the setting of extensive ischemia (> 20% myocardium), a 30% reduction in risk of all-cause death was present with the use of early revascularization compared with MT. Equipoise between the two strategies was present with approximately 10%-15% of the myocardium ischemic. As for patients with < 10% fixed defect, the risk reduction was 12.5% with MT and for patients with prior revascularization but no prior myocardial infarction it was 7.5%. Thus, these studies demonstrate that for patients with moderate to severe ischemia, revascularization is the preferred therapy for survival benefit, whereas for patients with no to mild ischemia MT is the main choice and revascularization is associated with increased mortality.

There is some evidence that revascularization in patients with no or mild ischemia is not associated with improved ischemia, but rather associated with worsened ischemia. Safley et al[38] identified 301 patients who underwent PCI for chronic total occlusion and in whom MPI was performed within 12 ± 3 mo before PCI and a follow-up study within 12 ± 3 mo after PCI. The change in % ischemia was +5.39% (P = 0.006), -1.70% (P = 0.008), -6.32% (P < 0.001), and -16.26% (P < 0.001) in patients with no/minimal (< 5% ischemic myocardium), mild (5%-9.9%), moderate (10%-16%), and severe (> 16%) ischemia, respectively. The percentage of patients with improved ischemic myocardium ≥ 5% was 0%, 34.7%, 68.5%, and 86.7% in patients with no/minimal, mild, moderate, and severe ischemia, respectively (P < 0.001). The percentage of patients with worsened ischemic myocardium ≥ 5% was 87.3%, 34.7%, 19.2%, and 9.2% in patients with no/minimal, mild, moderate, and severe ischemia, respectively (P < 0.001). Kaplan-Meier survival in patients with vs without improvement in ischemia showed a survival advantage in patients with improved ischemic myocardium ≥ 5% (87% vs 78%, P = 0.018). Receiver operating characteristics curve (ROC) analysis identified a 12.5% ischemic burden as the optimal cut-point to predict improvement in ischemia following PCI (sensitivity 80%, specificity 80%). This 12.5% ischemic burden is almost the same as that in the 2011 study by Hachamovitch et al[23]. Also ROC analysis identified a 6.25% ischemic burden as the optimal cut-point to predict worsening in ischemia following PCI (sensitivity 75%, specificity 80%). Thus, this study demonstrated that revascularization had no survival benefit and harms patients with no to mild ischemia, although the study was limited to patients who underwent PCI for chronic total occlusion.

Myocardial infarction associated with PCI (periprocedural myocardial infarction) is classified as type 4a by the third universal definition of myocardial infarction[39]. The prevalence of periprocedural myocardial infarction is 7.3% to 17.9% defined by CK-MB isoenzyme elevation > 3x upper limit of normal (ULN) and 15.0% to 44.2% defined by cardiac troponin > ULN[40-55]. The results of several studies suggested that any elevation in CK-MB was associated with reduced long-term survival and that there was a direct correlation between the magnitude of myonecrosis and mortality. Other studies have shown that only large myocardial infarctions were predictive of a poor long-term outcome[40-46]. Similarly, some studies showed that the serum concentration of cardiac troponin was an independent predictor of survival, others did not[47-55]. However two recent meta-analyses concluded that an elevated cardiac troponin levels after PCI does provide prognostic information[56,57]. Risk factors of periprocedural myocardial infarction are those which identify patients with increasing atherosclerotic disease burden, increased thrombotic risk, and with neurohormonal activation that predispose to either macrovascular complications (side branch occlusion or macroembolization) or microvascular obstruction (distal embolization of microparticles)[58].

In the era of coronary angioplasty, many studies reported that numerous “false positive” reversible perfusion defects occurred early after angioplasty, possibly as a result of inadequate early vessel remodeling or sustained abnormalities of coronary vasomotor tone. However, a significant percentage of patients showed persistent abnormalities in the later period[59-66]. In one study, 76% of patients without prior myocardial infarction showed improvement in perfusion abnormalities after angioplasty, but only 34% had completely reversible ischemia[60]. In the other study of 15 patients 1 to 2 wk after angioplasty, 7 had a reversible perfusion defect, of whom only 4 subsequently normalized by 4 to 6 wk[61]. These studies suggested that an improved or normalized perfusion abnormality does not necessarily occur after coronary angioplasty in every patient. Taken together, revascularization in patients with no or mild ischemia is likely to result in worsened ischemia, which is associated with increased mortality.

There are some studies which showed that the ischemia-guided (IG) strategy resulted in a better prognosis[67-70]. Farzaneh-Far et al[67] identified 1425 consecutive patients with coronary artery disease who underwent two serial MPI. They were followed for a median of 5.8 years after the second MPI. Patients were included in the PCI or coronary artery bypass graft (CABG) group on the basis of the first revascularization procedure occurring within 60 d of the first MPS scan. Thus patients were divided into a MT group, PCI group, and CABG group. The incidence of patients with worsening of the ischemic myocardium by ≥ 5% was more frequent in the MT group (15.6%) compared with the PCI (6.2%) and CABG groups (6.7%) (P < 0.001). After adjustment for established predictors, ≥ 5% ischemia worsening remained a significant independent predictor of death or myocardial infarction (HR = 1.634; P = 0.0019). Thus, this study showed that ischemia worsening was an independent predictor of death or myocardial infarction, and revascularization was associated with more frequent improvement in myocardial ischemia compared with MT.

Kim et al[68] studied the importance of IG revascularization. From a registry of 5340 patients with multivessel coronary artery disease, comprising 2587 PCI and 2753 CABG. MPI was performed in 42.3% of patients and IG revascularization was performed in 17.3%. The MPI was defined as abnormal if the SSS was 3 or greater. The incidence of major adverse cardiac and cerebrovascular events (MACCE) was significantly lower in the IG group than in the non-IG group [16.2% vs 20.7%, adjusted HR (aHR) = 0.73; 95%CI: 0.60-0.88; P = 0.001], primarily driven by the lower repeat revascularization rate (9.9% vs 22.8%, aHR = 0.66; 95%CI: 0.49-0.90; P = 0.009). Subgroup analysis showed that IG reduced the risk of MACCE in PCI patients (17.4% vs 22.8%, aHR = 0.59; 95%CI: 0.43-0.81; P = 0.001) but not in CABG patients (16.0% vs 18.5%, aHR = 0.87; 95%CI: 0.67-1.14; P = 0.31). Thus IG revascularization with MPI, particularly in PCI-treated patients, seems to decrease the risk of repeat revascularization and MACCE in patients with multivessel disease. Taken together, these studies suggest that the IG strategy is associated with improved prognosis.

FFR (the ratio of maximal blood flow in a stenotic artery to normal maximal flow), is now a gold standard for invasive assessment of coronary artery stenosis[71-80]. In Fractional Flow Reserve vs Angiography in Multivessel Evaluation (FAME) study, investigators randomly assigned 1005 patients with multivessel coronary artery disease to PCI with implantation of drug-eluting stents guided by angiography alone or guided by FFR measurements in addition to angiography[81]. Patients assigned to angiography-guided PCI underwent stenting of all indicated lesions, whereas those assigned to FFR-guided PCI underwent stenting of all indicated lesions only if the FFR was 0.80 or less. The primary endpoint was the rate of death, nonfatal myocardial infarction, and repeat revascularization at 1 year. The number of indicated lesions per patient was 2.7 ± 0.9 in the angiography group and 2.8 ± 1.0 in the FFR group (P = 0.34). The number of stents used per patient was 2.7 ± 1.2 and 1.9 ± 1.3, respectively (P < 0.001). The 1-year event rate was 18.3% in the angiography group and 13.2% in the FFR group (P = 0.02). The rate of death and myocardial infarction was 11.1% in the angiography group and 7.3% in the FFR group (P = 0.04). Pijls et al[82] reported the 2-year follow-up results of the FAME study. The 2-year rates of mortality or myocardial infarction were 12.9% in the angiography-guided group and 8.4% in the FFR-guided group (P = 0.02). Combined rates of death, nonfatal myocardial infarction, and revascularization were 22.4% and 17.9%, respectively (P = 0.08). For lesions deferred on the basis of FFR > 0.80, the rate of myocardial infarction was 0.2% and the rate of revascularization was 3.2% after 2 years, which is a very low rate. Thus, routine measurement of FFR in patients with multivessel coronary artery disease who undergo PCI with drug-eluting stents significantly reduced the rate of death, nonfatal myocardial infarction, and repeat revascularization for up to 2 years.

Tonino et al[83] studied the angiographic vs functional severity of coronary artery stenosis in the FAME study. Of the 1414 lesions (509 patients) in the FFR-guided arm of the FAME study, 1329 were successfully assessed by the FFR. Before FFR measurement, these lesions were categorized into 50%-70%, 71%-90%, and 91%-99% diameter stenosis by visual assessment. In the category 50%-70% stenosis, only 35% were functionally significant. In the category 71%-90% stenosis, 80% were functionally significant and in the category of subtotal stenoses, 96% were functionally significant. Of all 509 patients with angiographically defined multivessel disease, only 235 (46%) had functional multivessel disease.

In FAME 2 study, investigators enrolled patients with stable coronary artery disease for whom PCI was being considered, and assessed all stenoses by measuring FFR[84]. Patients in whom at least one stenosis was functionally significant (FFR ≤ 0.80) were randomly assigned to FFR-guided PCI plus the best available MT (PCI group), or the best available MT alone (MT group). Patients in whom all stenoses had an FFR of more than 0.80 were entered into a registry and received the best available MT. The primary endpoint was a composite of death, myocardial infarction, or urgent revascularization. Recruitment was halted prematurely after enrollment of 1220 patients (888 who underwent randomization and 332 enrolled in the registry) because of a significant between-group difference in the percentage of patients who had a primary endpoint event: 4.3% in the PCI group and 12.7% in the MT group (HR with PCI: 0.32; 95%CI: 0.19-0.53; P < 0.001). The difference was driven by a lower rate of urgent revascularization in the PCI group than in the MT group (1.6% vs 11.1%; HR = 0.13; 95%CI: 0.06-0.30; P < 0.001). Among patients in the registry, 3.0% had a primary endpoint event, which was not significantly different from the PCI group. Thus, in patients with stable coronary artery disease and functionally significant stenoses, FFR-guided PCI plus the best available MT, as compared with the best available MT alone, decreased the need for urgent revascularization. In patients without ischemia, the outcome appeared to be favorable with the best available MT alone. The main reason why there was no significant difference in death and myocardial infarction between the PCI group and MT group seems to be the relatively small number of patients and short-term follow-up period (mean duration of follow-up was 213 ± 128 d in the PCI group and 214 ± 127 d in the MT group).

Pijls et al[80] explain why FFR-guided PCI decreases the rate of death and myocardial infarction in the FAME study. From many studies it is known that the death and myocardial infarction rates are less than 1% per year for a functionally nonsignificant stenosis if treated appropriately by medication, between 5% and 10% per year for a functionally significant stenosis if only treated by medication, and approximately 3% per year for a stented lesion whether it was functionally significant or not. Thus, stenting a functionally significant stenosis improves outcome, but stenting a functionally nonsignificant stenosis worsens outcome. Taken together, these studies suggest that IG PCI is superior to angiography-guided PCI, and the presence of ischemia is the key to the decision-making for PCI.

For many years, the American College of Cardiology (ACC) and American Heart Association (AHA) have jointly published and updated guidelines for PCI and CABG[85,86]. Recently, the ACC Foundation/Society for Cardiovascular Angiography and Interventions/Society for Thoracic Surgeons/American Association for Thoracic Surgery/AHA/American Society of Nuclear Radiology released appropriateness criteria for coronary revascularization to serve as a supplement to the ACC/AHA guideline documents[87].

Hannan et al[88] studied the appropriateness of PCI and CABG performed in New York for patients without acute coronary syndrome or previous CABG. Of the 8168 patients undergoing CABG, 90.0% were appropriate for revascularization, 1.1% were inappropriate, and 8.6% were uncertain. Of the 33970 PCI patients, 28% lacked sufficient information to be rated. Of the patients who could be rated, 36.1% were appropriate, 14.3% were inappropriate, and 49.6% were uncertain. A total of 91% of the patients undergoing PCI who were classified as inappropriate had one- or two-vessel disease without proximal left anterior descending artery disease, and had no or minimal anti-ischemic MT. Chan et al[89] studied 500154 patients enrolled in the National Cardiovascular Data Registry. For 355417 patients with acute indications, 98.6% were classified as appropriate, 1.1% as inappropriate, and 0.3% as uncertain. For 144737 patients with nonacute indications, 50.4% were classified as appropriate, 11.6% as inappropriate, and 38.0% as uncertain. The majority of inappropriate PCIs for nonacute indications were performed in patients with no angina (53.8%), low-risk ischemia on noninvasive stress testing (71.6%), or suboptimal (≤ 1 medication) antianginal therapy (95.8%). Furthermore, although variation in the proportion of inappropriate PCI across hospitals was minimal for acute procedures, there was substantial hospital variation for nonacute procedures (mean hospital rate for inappropriate PCI, 10.8%; interquartile range, 6.0%-16.7%).

Lin et al[90] studied the frequency and predictors of stress testing prior to elective PCI in a Medicare population of 23887 patients. Only 44.5% of patients underwent stress testing within 90 d prior to elective PCI. There were wide regional variations among the hospital referral regions, with stress testing ranging from 22.1% to 70.6% (mean, 44.5%, interquartile range 39.0%-50.9%). Female sex [adjusted OR (aOR) = 0.91; 95%CI: 0.86-0.97], age 85 years or older (aOR = 0.83; 95%CI: 0.72-0.95), a history of congestive heart failure (aOR = 0.85; 95%CI: 0.79-0.92), and prior cardiac catheterization (aOR = 0.45; 95%CI: 0.38-0.54) were associated with a decreased likelihood of prior stress testing. Thus, these studies demonstrated that, although PCI in the acute setting and CABG are properly performed in most patients, PCI in the nonacute setting is often inappropriate, and stress testing to identify myocardial ischemia is performed in less than half of patients.

Some studies also showed that revascularization in an inappropriate setting is not associated with improved prognosis. Ko et al[91] assessed the appropriateness of coronary revascularization (PCI or CABG) and examined its association with longer-term outcomes. In 1625 patients with stable coronary artery disease, coronary revascularization was performed in only 69% in the appropriate category, 45% in the inappropriate category, and 54% in the uncertain category. In patients in the appropriate category, coronary revascularization was associated with a lower adjusted hazard of death or acute coronary syndrome (aHR = 0.61; 95%CI: 0.42-0.88; P = 0.0087) at 3 years compared with MT. No significant differences in death or acute coronary syndrome were observed between coronary revascularization and MT in the inappropriate category (aHR = 0.99; 95%CI: 0.48-2.02) and the uncertain category (aHR = 0.57; 95%CI: 0.28-1.16; P = 0.12).

Both MPI and FFR clearly identify the presence or absence of myocardial ischemia, and IG revascularization is associated with improved prognosis. However, the FFR value which is concordant with a 10% ischemic myocardium by MPI remains to be determined. A cut-off value of 0.75 was determined by the positive or negative results of three noninvasive stress tests; bicycle exercise test, thallium scintigraphy, and stress echocardiography with dobutamine[92]. A FFR value between 0.75 and 0.80 is deemed to be in the gray zone. MPI has limitation in identification of the highest risk subsets, left main coronary artery disease and three-vessel coronary artery disease, because of “balanced ischemia”[93-98]. One study showed that in patients with left main coronary artery disease, MPI results were normal in 5% and low-risk in 10% of patients[93]. The other study showed that in patients with triple-vessel coronary artery disease, MPI results were normal in 12% and single-vessel in 28% of patients[94].

Some studies compared MPI and FFR in patients with multivessel coronary artery disease. Ragosta et al[99] performed angiography, FFR, and MPI in 36 patients (88 arteries), and determined the association between FFR and perfusion for each vascular zone. Concordance between angiography, FFR, and MPI was seen in 61 of 88 zones (69%). Discordance was seen in the remaining 27 zones (31%), and was predominantly related to the finding of a FFR < 0.75 or total occlusion despite no defect on MPI. Melikian et al[100] performed MPI and FFR in 67 patients (201 vessels) with angiographic two- or three-vessel coronary artery disease. In 42% of patients, MPI and FFR detected identical ischemic areas (mean number of areas 0.9 ± 0.8 for both, P = 1.00). In the remaining 36% MPI underestimated the number (MPI = 0.46 ± 0.6, FFR = 2.0 ± 0.6, P < 0.001) and in 22% overestimated the number (MPI = 1.9 ± 0.8, FFR = 0.5 ± 0.8, P < 0.001) in comparison with FFR. Thus, MPI has poor concordance with FFR and tends to underestimate or overestimate the functional importance of coronary stenosis in comparison with FFR in patients with multivessel disease. In patients with multivessel coronary artery disease, FFR is the preferred method to identify myocardial ischemia. Therefore, complementary use of noninvasive MPI and invasive FFR would be important to compensate for each method’s limitations.

MPI studies demonstrate that for patients with moderate to severe ischemia, revascularization is the preferred therapy for survival benefit. For patients with no to mild ischemia, MT is the main choice and revascularization is associated with increased mortality probably because of worsened ischemia. FFR studies demonstrate that IG PCI is superior to angiography-guided PCI, and the presence of ischemia is the key to decision-making for PCI. Studies of appropriateness criteria demonstrate that, although CABG and emergency PCI are appropriately performed in most patients, use of elective PCI is often inappropriate. Some studies also suggest that revascularization in an inappropriate setting is not associated with improved prognosis. Taken together, myocardial ischemia is a key factor in the management of patients with stable coronary artery disease.