American College of Cardiology

A broad spectrum of clinical presentations exists wherein patients may be considered candidates for PCI, ranging from asymptomatic to severely symptomatic or unstable, with variable degrees of jeopardized myocardium. Selection of appropriate candidates for PCI in a variety of clinical presentations is reviewed in this section.

Each time that a patient is considered for revascularization, the potential risk and benefits of the particular procedure under consideration must be weighed against alternative therapies (Table 18).

When PCI is considered, the benefits and risks of surgical revascularization and medical therapy always deserve thoughtful discussion with the patient and family. The initial simplicity and associated low morbidity of PCI as compared to surgical therapy is always attractive, but the patient and family must understand the limitations inherent in current PCI procedures, including a realistic presentation of the likelihood of restenosis and the potential for incomplete revascularization as compared with CABG surgery. In patients with CAD who are asymptomatic or have only mild symptoms, the potential benefit of antianginal drug therapy along with an aggressive program of risk reduction must also be understood by the patient before a revascularization procedure is performed.

A. Asymptomatic or Mild Angina

In the previous ACC/AHA Guidelines for PTCA, specific recommendations were made separately for patients with single- or multivessel disease ^(16,98). The current techniques of PCI have matured to the point where, in patients with favorable anatomy, the competent practitioner can perform either single- or multivessel PCI at low risk and with a high likelihood of initial success. For this reason, in this revision of the Guidelines, recommendations will be made largely based upon the patients' clinical condition, specific coronary lesion morphology and anatomy, LV function, and associated medical conditions, and less emphasis will be placed on the number of lesions or vessels requiring PCI. The CCS Class of angina (I to IV) is used to define the severity of symptoms. The categories described in this section refer to an initial PCI procedure in a patient without prior CABG surgery. The randomized trials comparing PTCA and medical therapy have been discussed (Table 11).

The Committee recognizes that the majority of patients with asymptomatic ischemia or mild angina should be treated medically. The published ACIP study ⁽¹⁷⁶⁾ casts some doubt on the wisdom of medical management for those higher-risk patients who are asymptomatic or have mild angina, but have objective evidence by both treadmill testing and ambulatory monintoring of significant myocardial ischemia and CAD. In addition, there is a substantial portion of the middle and older age populations in this country that remains physically active, participating in sports, such as tennis and skiing, or performing regular and vigorous physical exercise, such as jogging, who have CAD. For such individuals with moderate or severe ischemia and few symptoms, revascularization with PCI or CABG surgery may reduce their risk of serious or fatal cardiac events. For this reason, patients in this category of higher-risk asymptomatic ischemia or mild symptoms and severe anatomic CAD are placed in Class I or II. PCI may be considered if there is a high likelihood of success and a low risk of morbidity or mortality. The judgment of the experienced physician is deemed valuable in assessing the extent of ischemia.

Recommendations for PCI in Asymptomatic or Class I Angina Patients (Table 19)

Class I
1. Patients who do not have treated diabetes with asymptomatic ischemia or mild angina with 1 or more significant lesions in 1 or 2 coronary arteries suitable for PCI with a high likelihood of success and a low risk of morbidity and mortality. The vessels to be dilated must subtend a large area of viable myocardium ⁽¹⁰⁸⁾ (Table 20). (Level of Evidence: B)

Class IIa
1. The same clinical and anatomic requirements for Class I, except the myocardial area at risk is of moderate size or the patient has treated diabetes. (Level of Evidence: B)

Class IIb
1. Patients with asymptomatic ischemia or mild angina with 3 coronary arteries suitable for PCI with a high likelihood of success and a low risk of morbidity and mortality. The vessels to be dilated must subtend at least a moderate area of viable myocardium. In the physician’s judgment, there should be evidence of myocardial ischemia by ECG exercise testing, stress nuclear imaging, stress echocardiography or ambulatory ECG monitoring or intracoronary physiologic measurements. (Level of Evidence: B)

Class III
1. Patients with asymptomatic ischemia or mild angina who do not meet the criteria as listed under Class I or Class II and who have:

a. Only a small area of viable myocardium at risk
b. No objective evidence of ischemia
c. Lesions that have a low likelihood of successful dilatation
d. Mild symptoms that are unlikely to be due to myocardial ischemia
e. Factors associated with increased risk of morbidity or mortality
f. Left main disease
g. Insignificant disease <50% (Level of Evidence: C)

B. Angina Class II to IV or Unstable Angina

Many patients with moderate or severe stable angina or unstable angina do not respond adequately to medical therapy and often have significant coronary artery stenoses that are suitable for revascularization with CABG surgery or PCI. In addition, a proportion of these patients have reduced LV systolic function, which places them in a group that is known to have improved survival with CABG surgery and possibly with revascularization by PCI ^{(178,179,240,241)}. In nondiabetic patients with 1- or 2-vessel disease in whom angioplasty of 1 or more lesions has a high likelihood of initial success, PCI is the preferred approach. In a minority of such patients, CABG surgery may be preferred, particularly for those in whom the left anterior descending coronary artery can be revascularized with the internal mammary artery or in those with left main coronary disease. In patients with unstable angina or non-Q-wave MI, intensive medical therapy should be initiated prior to revascularization with PCI or CABG surgery ^(242-244).

Clinical investigations evaluating the use of routine catheterization and PCI for patients with unstable angina and NSTEMI (non-ST-segment elevation myocardial infarction) have yielded inconsistent results. TIMI-IIIB was the first to compare strategies of routine catheterization and revascularization in addition to medical therapy and selective use of aggressive treatment. In TIMI-IIIB, there was no difference in the incidence of death or recurrent MI at 1 year between the 2 strategies, but patients treated by the aggressive strategy experienced less angina and repeat hospitalizations for ischemia and required fewer medications ⁽²⁴⁵⁾. In the VANQWISH trial performed by the Veterans Administration, no difference in death or death and MI was observed between the 2 strategies at late follow-up, but the minority of patients in the aggressive strategy received revascularization, and the mortality rate for those having CABG was high ⁽²⁴⁶⁾. The FRISC II trial compared medical and revascularization approaches among patients after 6 days of low molecular weight heparin therapy before a decision regarding PCI ⁽²⁴⁷⁾. Those randomized to the conservative therapy only underwent PCI if they had 3 mm ST depression on stress testing. Compared with prior studies, patients assigned to the aggressive strategy in FRISC II experienced a 22% reduction (p = 0.031) in the incidence of death or MI at 6 months (9.4%) compared to conservatively treated patients (12.1%). In addition, there was a significant decrease in myocardial infarction rate alone and a non-significantly lower mortality rate in the treated group (1.9% vs. 2.9%; p = 0.10). Symptoms of angina and hospital readmission were decreased 50% by the invasive strategy. These findings were supported by long-term follow-up from the FRISC II study indicating that low-molecular-weight heparin and early intervention lowered the risk of death, myocardial infarction, and revascularization in unstable coronary syndromes, at least during the first 1 month of therapy. Early protective therapy could be used to lower the risk of late events in patients waiting for definitive PCI ⁽²⁴⁸⁾. This treatment benefit was most pronounced for high-risk patients. The FRISC II trial ⁽²⁴⁷⁾ results support the use of catheterization and revascularization for selected patients with an acute coronary syndrome. The Treat Angina with Aggrastat and determine the Cost of Therapy with an Invasive or Conservative Strategy (TACTICS) Trial randomized 2220 patients to an early invasive strategy in which cardiac catheterization and revascularization were performed 4 to 48 h after randomization or to a conservative strategy in which revascularization was reserved for those patients who developed recurrent ischemia after medical stabilization. All patients were treated with aspirin, heparin, ß-blockers, cholesterol-lowering therapy, and tirofiban. The primary endpoint, a composite of death, MI, and rehospitalization for worsening chest pain by 6 months, was lower in patients assigned to the invasive strategy (15.9% vs. 19.4% in patients assigned to conservative therapy; p = 0.0025). The rate of death or MI was also significantly reduced at 6 months in the invasive strategy arm (7.3% vs. 9.5% in patients assigned to conservative therapy; p < 0.05) ⁽²⁴⁹⁾. These promising results have not yet undergone peer review and have not been published.

The indications for coronary angiography are summarized in the ACC/AHA Coronary Angiography Guidelines ⁽¹⁹⁴⁾, and recommendations for PCI are summarized in the ACC/AHA Unstable Angina Guidelines ⁽²⁵⁰⁾. Indications for PCI for patients with angina Class II to IV, unstable angina, or non-Q-wave infarction follow.

Recommendations for Patients with Moderate or Severe Symptoms (Angina Class II to IV, Unstable Angina or Non-ST-Elevation MI) With Single- or Multivessel Coronary Disease on Medical Therapy (Table 21)

Class I
1. Patients with 1 or more significant lesions in 1 or more coronary arteries suitable for PCI with a high likelihood of success and low risk of morbidity or mortality (Tables 6 and 8). The vessel(s) to be dilated must subtend a moderate or large area of viable myocardium and have high risk (Table 20). (Level of Evidence: B)

Class IIa
1. Patients with focal saphenous vein graft lesions or multiple stenoses who are poor candidates for reoperative surgery. (Level of Evidence: C)

Class IIb
1. Patient has 1 or more lesions to be dilated with reduced likelihood of success (Table 5) or the vessel(s) subtend a less than moderate area of viable myocardium. Patients with 2- or 3-vessel disease, with significant proximal LAD CAD and treated diabetes or abnormal LV function. (Level of Evidence: B)

Class III
1. Patient has no evidence of myocardial injury or ischemia on objective testing and has not had a trial of medical therapy, or has

a. Only a small area of myocardium at risk
b. All lesions or the culprit lesion to be dilated with morphology with a low likelihood of success
c. A high risk of procedure-related morbidity or mortality. (Level of Evidence: C)

2. Patients with insignificant coronary stenosis (e.g., <50% diameter). (Level of Evidence: C)

3. Patients with significant left main CAD who are candidates for CABG. (Level of Evidence: B)

It is recognized by the Committee that the assessment of risk of unsuccessful PCI or serious morbidity or mortality must always be made with consideration of the alternative therapies available for the patient, including more intensive or prolonged medical therapy or surgical revascularization (Table 22), especially in patients with unstable angina pectoris.

When CABG surgery is a poor option because of high risk due to special considerations or other organ system disease, patients otherwise in Class IIb may be appropriately managed with PCI. Under these special circumstances formal surgical consultation is recommended.

C. Myocardial Infarction

The results of randomized clinical trials of intravenous thrombolysis and subsequent management strategies of immediate, delayed, and deferred PCI have established the benefits of early pharmacologic and mechanical reperfusion therapies for patients with acute MI ^{(209,210,251-256)}.

Acute MI results from a severe and sudden cessation of myocardial blood flow, most commonly due to atherosclerotic-thrombotic occlusion of a major epicardial coronary artery. PCI is a very effective method for re-establishing coronary perfusion and is suitable for 90% of patients. Considerable data support the use of PCI for patients with acute MI ^(257,258). Reported rates of achieving TIMI 3 flow, the goal of reperfusion therapy, range from 70 to 90% ⁽²⁵⁹⁾. Late follow-up angiography demonstrates that 87% of infarct arteries remain patent ⁽²⁶⁰⁾. Although most evaluations of PCI have been in patients who are eligible to receive thrombolytic therapy, considerable experience supports the value of PCI for patients who may not be suitable for thrombolytic therapy due to an increased risk of bleeding ⁽²⁶¹⁾.

Intracoronary stents appear to augment the results of PCI for MI (Table 23). Preliminary results suggest that stenting achieves a better immediate angiographic result with a larger arterial lumen, less reclosure of the infarct-related artery, and fewer subsequent ischemic events than PTCA alone ^(262-264). Results from a randomized clinical trial suggest that stenting enhances late clinical outcomes (reduction in composite endpoint attributable to a decrease in target vessel revascularization) when compared to PTCA alone ⁽²⁶⁴⁾. However, an increase in mortality at 1 year among the stent group has been reported in the Stent-PAMI trial ⁽²⁶⁵⁾.

Primary PTCA performed without routine stenting has been compared to thrombolytic therapy in several randomized clinical trials. These investigations consistently demonstrate that PTCA-treated patients experience less recurrent ischemia or infarction than those treated by thrombolysis ^(266-269). Trends favoring a survival benefit with PTCA are noted. The most recent and largest single trial (1138 patients) demonstrated significant benefit in the composite endpoint death, recurrent MI, or disabling stroke at 30 days favoring angioplasty, although this benefit was not sustained at 6 months ^(260,270). 2 Meta-analyses showed superiority of PCI over thrombolysis for mortality with risk reductions of 0.34 and 0.56 ^(271,272). It is important to note that these results of PCI have been achieved in medical centers with experienced providers and under circumstances where angioplasty can be performed immediately following patient presentation (Figure 3).

1. PCI in Thrombolytic-Ineligible Patients. Randomized, controlled clinical trials evaluating the outcome of PCI for patients who present with ST-segment elevation but who are ineligible for thrombolytic therapy and for patients who experience infarction without ST-segment elevation have not been performed. Nevertheless, there is a general consensus that PCI is an appropriate means for achieving reperfusion in patients who cannot receive thrombolytics because of increased risk of hemorrhage. Other reasons also exclude acute MI patients from thrombolytic therapy, and the outcome of PCI in these patients may differ from those eligible for lytic therapy. For example, patients who present without ST-elevation are more often older and female and have higher in-hospital mortality than those with ST-segment elevation. Little data are available to characterize the value of primary PCI for this subset of acute MI patients ⁽²⁶¹⁾ (Table 24).

2. Post-Thrombolysis PCI. In asymptomatic patients, the strategies of routine PCI of the stenotic infarct-related artery immediately after successful thrombolysis show no benefit with regard to salvage of jeopardized myocardium or prevention of reinfarction or death. In some studies this approach was associated with increased incidence of adverse events, which include bleeding, recurrent ischemia, emergency coronary artery surgery, and death ^(279-282). Routine PCI immediately after thrombolysis may increase the chance for vascular complications at the catheterization access site and hemorrhage into the infarct-related vessel wall ⁽²⁸²⁾.

Spontaneous recurrent ischemia and reinfarction have been observed to occur in approximately 15 to 25% of thrombolytic-treated patients ^{(131,254,283)}. The majority of spontaneous cardiac ischemic events occur within the first 24 to 48 h following treatment with thrombolytic therapy and are associated with an increase in-hospital morbidity and mortality ^(284-286). Patients at risk for recurrent ischemia tend to be older and have more anterior infarcts. Some thrombolytic-treated patients initially managed conservatively will require urgent cardiac catheterization and revascularization because of recurrent MI ^(287,288).

To assess whether low-dose alteplase with standard dose abciximab enhanced 90-min reperfusion after acute MI, the strategies for patency enhancement in the emergency department (SPEED study group) examined the outcomes of 484 patients divided into 5 groups receiving combinations of abciximab, with and without low-dose reteplase, reteplase alone, and standard reteplase. The results of this trial indicated that adding reteplase to abciximab treatment for acute MI vs. reteplase alone enhanced the incidence of early complete reperfusion after initiation of therapy in the emergency department ⁽²⁸⁷⁾. Similar data have been supportive from the GUSTO-IV trial and GUSTO-I investigators ^(288,289).

3. Rescue PCI. Rescue (also known as salvage) PCI is defined as PCI after failed thrombolysis for patients with continuing or recurrent myocardial ischemia. Rescue PCI has resulted in higher rates of early infarct-artery patency, improved regional infarct zone wall motion, and greater freedom from adverse in-hospital clinical events compared to a deferred PCI strategy ⁽²⁹⁰⁾. The randomized evaluation of rescue PCI with combined utilization endpoints trial (RESCUE) demonstrated a reduction in rates of in-hospital death and combined death and congestive heart failure maintained up to 1 year after study entry for patients presenting with anterior wall MI who failed thrombolytic therapy ^(291,292). Improvement in TIMI grade flow from 2 to 3 may offer additional clinical benefit.

4. PCI for Cardiogenic Shock. Observational studies support the value of PCI for patients who develop cardiogenic shock in the early hours of MI. For patients who do not have mechanical causes of shock, such as acute mitral regurgitation or septal or free wall rupture, mortality among those having PCI is lower than those treated by medical means ⁽²²²⁾. However, having catheterization alone, with or without angioplasty, is associated with a low mortality ^(222,293). Thus, the relatively favorable outcome of angioplasty-treated patients may be, in part, due to a bias in patient selection. Since the outcome of cardiogenic shock is so unfavorable with contemporary medical therapy, angioplasty continues to be recommended as potential life-saving therapy ⁽²²²⁾.

A randomized clinical trial has further clarified the role of emergency revascularization in acute MI complicated by cardiogenic shock ⁽²²²⁾. In this study, 302 patients with acute MI and cardiogenic shock were randomly assigned to emergency revascularization (ERV, n = 152) by coronary angioplasty or bypass surgery or to initial medical stabilization (IMS, n = 150). The 30-day mortality was significantly lower (p 0.01) for patients <75 years old treated with ERV (41.1% mortality) compared to IMS (56.8% mortality). By contrast, mortality among patients >75 years was worse for those treated with ERV. The use of an IABP was the same in both groups (86%). Among those receiving ERV, 60% had PTCA and 40% received CABG with 30-day mortality rates of 45% and 42% respectively. For the IMS group, 63% received thrombolytic agents and 25% had delayed revascularization. This multicenter trial supports the use of ERV with PCI in appropriate candidates for patients <75 years old with acute MI complicated by cardiogenic shock ⁽²²²⁾. Hochman et al. ⁽²²²⁾ also demonstrated that in patients with cardiogenic shock, emergency revascularization did not significantly reduce overall mortality at 30 days. However, after 6 months, there was significant survival benefit to early revascularization. These data strongly support the approach that patients less than age 75 with acute MI complicated by cardiogenic shock should undergo emergency revascularization and support measures.

5. PCI Hours to Days After Thrombolysis. Patients who achieve reperfusion and myocardial salvage following thrombolytic therapy may experience reocclusion of the infarct artery and recurrent MI. This concern has prompted the routine use of catheterization and PCI prior to hospital discharge to identify and dilate the culprit lesion. The value of this approach was tested in 2 large, randomized clinical trials. The SWIFT study ⁽²⁸⁰⁾ examined 800 patients with acute MI randomly assigned to PCI within 2 to 7 days after thrombolysis or to conservative management with intervention for spontaneous or provokable ischemia. There were no differences in the 2 treatment strategies regarding LV function, incidence of reinfarction, in-hospital survival, or 1-year survival rate. Similarly in the TIMI-IIB trial ⁽²⁸¹⁾, 3262 patients randomized to angioplasty within 18 to 48 h vs. conservative management after acute infarct having received t-PA were examined. The 2 groups had similar mortality at 6 weeks (5.2% vs. 4.7%), incidence of nonfatal reinfarction (6.4% vs. 5.8%), and LV ejection fraction (0.5% vs. 0.5%). The 1- and 3-year survival rate, anginal class, and frequency of bypass surgery were also similar between the 2 groups ^{(131, 294)}. These data indicate that routine PCI of the infarct-related artery in the absence of spontaneous or provoked ischemia is not warranted.

A recent randomized trial comparing primary angioplasty with a strategy of short-acting thrombolysis and immediate planned rescue angioplasty in acute MI was performed by the PACT (Plasminogen-activator Angioplasty Compatibility Trial) investigators by Ross et al. ⁽²⁹⁵⁾. This study evaluated the safety and efficacy of reduced-dose fibrinolytic therapy to promote early infarct patency coupled with PCI. In 606 patients, 50 mg of rt-PA followed by immediate angioplasty was performed to recanalize infarct-related arteries. Endpoints of time-to-artery patency and technical results of angioplasty were reported. Patency in the catheterization laboratory on arrival was 61% with rt-PA and 34% with placebo. Rescue and primary angioplasty restored TIMI flow equally in both groups. There was no difference in the incidence of stroke or bleeding. Left ventricular function was highest in the patent infarct-related artery group on arrival to the catheterization laboratory. In 88% of angioplasty patients, the delay exceeded 1 h with a convalescent ejection fraction of 57%. These findings indicated that tailored thrombolytic regimes compatible with subsequent interventions lead to more frequent early recanalization (i.e., before arrival in the catheterization laboratory) which facilitates greater left ventricular preservation with no augmentation of adverse events at follow-up.

Initial studies of late (>6 to 12 h) PCI in asymptomatic survivors of MI, indicate that opening an occluded artery does not appear to alter the process of LV dilatation ⁽²⁹⁶⁾, the incidence of spontaneous and inducible arrhythmias ⁽²⁹⁷⁾, or prognosis ⁽²⁹⁸⁾. The TAMI-6 study ⁽²⁹⁹⁾ of angioplasty of a persistently occluded infarct artery 7 to 48 h after symptom onset demonstrated that the infarct-related artery patency was similar in aggressive or conservatively treated groups at 6-month follow-up. It was also noted at the end of the same 6-month follow-up that there was a high incidence of infarct-related artery patency in patients who did not receive angioplasty as well as a high incidence of reocclusion in those who did. LV ejection fraction, incidence of reinfarction, hospital admission, and mortality during follow-up were also similar between groups. In other similar studies with small patient numbers, late angioplasty of occluded infarct arteries improved LV performance, but convincing outcome data are lacking to support late angioplasty in asymptomatic patients within 48 h of failed thrombolysis ⁽³⁰⁰⁾.

The benefits of early reperfusion therapy, whether by thrombolytic drugs or PCI, have been attributed to salvage of severely ischemic myocardium, thereby limiting infarct size and preserving LV function. However, there is increasing evidence that achieving patency in the infarct-related artery, even hours to days after the acute event, may favorably influence the outcome by mechanisms other than myocardial salvage ^(301-303). Late restoration of patency appears to reduce infarct expansion ⁽²⁹⁶⁾ and ventricular remodeling ^(304,305), and attenuate the risk for the development of ventricular arrhythmias ^(297,306). These effects could all contribute to improvements in survival independent of the acute salvage of myocardium ^{(298,307,308)}. Although data supporting the argument to open occluded infarct-related arteries are persuasive, at least for large arteries subtending large areas of myocardium, there are few randomized trials supporting this approach. It should be noted that the overwhelming majority of trials were performed prior to the widespread use of stents and platelet IIb/IIIa receptor blockade and thus, the potential impact and benefit of these newer therapies in this clinical setting needs re-evaluation.

6. PCI After Thrombolysis in Selected Patient Subgroups

a. Young and Elderly Post-Infarct Patients. Although not supported by randomized trials, routine cardiac catheterization following thrombolytic therapy for AMI has been a frequently performed strategy in all age groups. Young (<50 years) patients often undergo cardiac catheterization after thrombolytic therapy due to a “perceived need” to define coronary anatomy and thus establish psychological as well as clinical outcomes. In contrast, older (>75 years) patients have higher in-hospital and long-term mortality rates and enhanced clinical outcomes when treated with primary PCI ⁽³⁰⁹⁾.

In a secondary analysis of the TIMI-IIB study comparing angiographic findings and clinical outcomes among 841 young (<50 years) and 859 older (65-70 years) patients randomly assigned to an invasive or conservative post-lytic management strategy ⁽³¹⁰⁾, the younger patients assigned to the invasive strategy commonly had insignificant (i.e., <60% diameter stenosis) and single-vessel CAD. Severe 3-vessel or left main coronary disease findings were infrequent (3-vessel incidence, 4%; left main, 0%). Fatal and nonfatal MI and death throughout the first year following study entry was also infrequent. There were no differences in the rates of in-hospital recurrent ischemia, reinfarction, or death among patients assigned to the conservative strategy of selective cardiac angiography and coronary revascularization as compared to an invasive strategy, consisting of routine post-lytic coronary angiography. Compared to younger patients, older patients had a higher prevalence of multivessel CAD (i.e., 44%) and high 42-day rates of reinfarction and death.

In spite of these observations, there was no difference in the 42-day rates of reinfarction or death among the older patient subgroup, regardless of the post-lytic management strategy. The TIMI-II data of younger and older infarct patients are consistent with the overall results of other randomized trials of thrombolysis/PTCA. Confirmatory studies to determine quality-of-life aspects of care in younger patients and to define the potential of other modes of coronary revascularization in older patient groups are not yet available. Based on the current data, with the exception of patients presenting with cardiogenic shock, use of PCI should be determined by clinical need without special consideration of age.

b. Patients With Prior Myocardial Infarction. A prior MI is an independent predictor of death, reinfarction, and need for urgent coronary bypass surgery ⁽³¹¹⁾. In thrombolytic trials, 14 to 20% of enrolled patients had a history of prior MI ^(254,281).

In the TIMI-II study, patients with a history of prior MI had a higher 42-day mortality (8.8% vs. 4.3%; p < 0.001), higher prevalence of multivessel CAD (60% vs. 28%; p < 0.001), and a lower LV ejection fraction (42% vs. 48%; p < 0.001) compared to patients with a first MI ⁽³¹²⁾. Among patients assigned to the conservative post-lytic strategy, those with a prior MI had a significantly higher 42-day mortality compared to patients with a first MI (11.5% vs. 3.5%; p < 0.001), whereas in the invasive strategy, the mortality outcome was essentially the same in the 2 patient groups. Mortality tended to be lower among patients with a prior MI undergoing the invasive compared to the conservative strategy, a benefit which persisted up to 1 year following study entry ⁽²⁹⁴⁾.

Based on the above findings and current practice, PCI should be based on clinical need. The presence of prior MI places the patient in a higher risk subset and should be considered in the PCI decision.

Recommendations for Primary PCI for Acute Transmural MI Patients as an Alternative to Thrombolysis (Table 25)

Class I
1. As an alternative to thrombolytic therapy in patients with AMI and ST-segment elevation or new or presumed new left bundle branch block who can undergo angioplasty of the infarct artery 12 h from the onset of ischemic symptoms or >12 h if symptoms persist, if performed in a timely fashion* by individuals skilled in the procedure†  and supported by experienced personnel in an appropriate laboratory environment.‡ (Level of Evidence: A)

2. In patients who are within 36 h of an acute ST elevation/Q-wave or new left bundle branch block MI who develop cardiogenic shock, are <75 years of age, and revascularization can be performed within 18 h of the onset of shock by individuals skilled in the procedure† and supported by experienced personnel in an appropriate laboratory environment.‡ (Level of Evidence: A)

* Performance standard: balloon inflation within 90 ± 30 min of hospital admission.
† Individuals who perform 75 PCI procedures per year.
‡ Centers that perform >200 PCI procedures per year and have cardiac surgical capability ^(193,194).

Class IIa
1. As a reperfusion strategy in candidates who have a contraindication to thrombolytic therapy. (Level of Evidence: C)

Class III
1. Elective PCI of a non-infarct-related artery at the time of acute MI. (Level of Evidence: C)

2. In patients with acute MI who:

a. have received fibrinolytic therapy within 12 h and have no symptoms of myocardial ischemia
b. are eligible for thrombolytic therapy and are undergoing primary angioplasty by an inexperienced operator (Individual who performs <75 PCI procedures per year)
c. are beyond 12 h after onset of symptoms and have no evidence of myocardial ischemia. (Level of Evidence: C)

Recommendations for PCI After Thrombolysis (Table 26)

Class I
1. Objective evidence for recurrent infarction or ischemia (rescue PCI) ⁽¹⁹⁴⁾. (Level of Evidence: B)

Class IIa
1. Cardiogenic shock or hemodynamic instability. (Level of Evidence: B)

Class IIb
1. Recurrent angina without objective evidence of ischemia/infarction. (Level of Evidence: C)

2. Angioplasty of the infarct-related artery stenosis within hours to days (48 h) following successful thrombolytic therapy in asymptomatic patients without clinical and/or inducible evidence of ischemia. (Level of Evidence: B)

Class III
1. Routine PCI within 48 h following failed thrombolysis. (Level of Evidence: B)

2. Routine PCI of the infarct-artery stenosis immediately after thrombolytic therapy. (Level of Evidence: A)

Recommendations for PCI During Subsequent Hospital Management After Acute Therapy for AMI Including Primary PCI (Table 27)

Class I
1. Spontaneous or provocable myocardial ischemia during recovery from infarction ⁽¹⁹⁴⁾. (Level of Evidence: C)

2. Persistent hemodynamic instability. (Level of Evidence: C)

Class IIa
1. Patients with LV ejection fraction 0.4, CHF, or serious ventricular arrhythmias. (Level of Evidence: C)

Class IIb
1. Coronary angiography and angioplasty for an occluded infarct-related artery in an otherwise stable patient to revascularize that artery (open artery hypothesis). (Level of Evidence: C)

2. All patients after a non-Q-wave MI. (Level of Evidence: C)

3. Clinical HF during the acute episode, but subsequent demonstration of preserved LV function (LV ejection fraction >0.4). (Level of Evidence: C)

Class III
1. PCI of the infarct-related artery within 48 to 72 h after thrombolytic therapy without evidence of spontaneous or provocable ischemia. (Level of Evidence: C)

D. Percutaneous Intervention in Patients With Prior Coronary Bypass Surgery

Ischemic symptoms recur in 4 to 8% of patients per year following CABG ^(313-316). Recurrence of symptoms can be attributed to progression of native vessel coronary disease (5% per year) and bypass conduit occlusion, particularly SVG failure (7% in week 1; 15 to 20% in first year; 1 to 2% per year during the first 5 to 6 years, and 3 to 5% per year in years 6 to 10 postoperatively) ^(313-315). At 10-years postoperatively, approximately half of all SVG conduits are occluded and only half of the remaining patent grafts are free of significant disease ^{(160,317-327)}. The requirement for repeat revascularization procedures increases over time from the initial revascularization, particularly in younger patients ⁽³²⁸⁾. Although arterial conduits exhibit improved long-term patency ^{(157,329,330)}, stenosis or occlusion of these grafts can occur. Thus, patients with recurrent ischemic symptoms following CABG may require repeat revascularization due to diverse anatomic problems.

Risk of repeat surgical revascularization is higher (hospital mortality 7 to 10%) than initial CABG ^(331-333) and both long-term relief of angina and bypass graft patency are lower than that of the first procedure ^{(331,334,335)}. In addition, patients with prior bypass surgery may have limited graft conduits, impaired LV function, advanced age, and coexisting medical conditions (cerebrovascular disease; renal and pulmonary insufficiency) which may complicate repeat surgical coronary revascularization and prompt consideration for catheter-based intervention. Patients with prior bypass surgery represent an increasing proportion of patients being referred for percutaneous coronary revascularization, and specific indications for therapy may be influenced by both anatomic considerations and the timing of recurrent ischemia postoperatively.

1. Early Ischemia After CABG. Recurrent ischemia early (<30 days) postoperatively usually reflects graft failure, often secondary to thrombosis ^(336-338), and may occur in both saphenous vein and arterial graft conduits ⁽³³⁹⁾. Incomplete revascularization and unbypassed native vessel stenoses or stenoses distal to a bypass graft anastomosis may also precipitate recurrent ischemia. Urgent coronary angiography is indicated to define the anatomic cause of ischemia and to determine the best course of therapy. Emergency PCI of a focal graft stenosis (venous or arterial) or recanalization of an acute graft thrombosis may successfully relieve ischemia in the majority of patients. Balloon dilatation across suture lines has been accomplished safely within days of surgery ^(340-342). Intracoronary thrombolytic therapy should be administered with caution during the first week postoperatively ^(343-346) and if required, residual thrombus may be “targeted” in low doses through a local drug delivery system. Conversely, mechanical thrombectomy with newer catheter technologies may be effective without the attendant risk of fibrinolysis ⁽³⁴⁷⁾. Adjunctive therapy with abciximab for percutaneous intervention during the first week following bypass surgery has been limited but intuitively may pose less risk for hemorrhage than fibrinolysis. As flow in vein graft conduits is pressure dependent, intra-aortic balloon pump support should be considered in the context of systemic hypotension and/or severe LV dysfunction. If feasible, PCI of both bypass graft and native vessel offending stenoses should be attempted, particularly if intracoronary stents can be successfully deployed.

When ischemia occurs 1 to 12 months following surgery, the etiology is usually peri-anastomotic graft stenosis. Distal anastomotic stenoses (both arterial and venous) respond well to balloon dilatation alone and have a more favorable long-term prognosis than stenoses involving the mid-shaft or proximal vein graft anastomosis ^{(135,136,348-351)}. Mid-shaft vein graft stenoses occurring during this time frame are usually due to intimal hyperplasia. Restenosis may be less frequent and event-free survival-enhanced following angioplasty of SVGs dilated within 6 months of surgery compared with grafts of older age. The immediate results of PCI in mid-shaft ostial or distal anastomotic vein graft stenoses may be enhanced by coronary stent deployment ^(351,352). Ablative technologies such as directional atherectomy or excimer laser coronary angioplasty may facilitate angioplasty and stent deployment in patients with aorto-ostial vein graft stenoses ^(353,354).

Stenoses in the mid-portion or origin of the internal mammary artery graft are uncommon, but respond to balloon dilatation ^(355,356) with stent deployment as feasible. Long-term follow-up of patients after internal mammary artery angioplasty has demonstrated sustained benefit and relief of ischemia in the majority of patients ^(357,358). Balloon angioplasty with or without stent deployment can be successfully performed in patients with distal anastomotic stenoses involving the gastroepiploic artery bypass graft and in patients with free radial artery bypass grafts as well ⁽³⁵⁹⁾. Percutaneous intervention has also been effective in relieving ischemia for patients with the stenosis of the subclavian artery proximal to the origin of a patent left internal mammary artery bypass graft ^(360,361).

2. Late Ischemia After CABG. Ischemia occurring more than 1 year postoperatively usually reflects the development of new stenoses in graft conduits and/or native vessels that may be amenable to PCI ⁽³⁶²⁾. At 3 years or more following SVG implantation, atherosclerotic plaque is frequently evident and is often progressive. These lesions may be friable and often have associated thrombus formation, which may contribute to the occurrence of slow flow, distal embolization, and periprocedural MI following attempted percutaneous intervention ⁽³⁶³⁾. Slow flow occurs more frequently in grafts having diffuse atherosclerotic involvement, angiographically demonstrable thrombus, irregular or ulcerative lesion surfaces, and with long lesions having large plaque volume ^(364,365).Although a reduced incidence of distal embolization has been reported following the use of the extraction atherectomy catheter to recanalize stenoses in older vein graft conduits ^(366-370), embolization may still complicate adjunctive balloon dilatation. Slow-flow with signs and symptoms of myocardial ischemia may be ameliorated by the intragraft administration of verapamil or diltiazem ^(364,371). The adjunctive administration of abciximab during vein graft intervention may reduce the incidence of distal embolization and non-Q-wave MI ⁽³⁷²⁾, but controversy remains regarding the benefit of prophylactic abciximab therapy in patients with prior coronary bypass surgery undergoing percutaneous intervention.

Although postprocedural minimum lumen diameter is larger following directional coronary atherectomy ^{(140,373,374)} or stent deployment ^{(139,141,142,375-381)} compared with balloon angioplasty of SVG stenoses, long-term prognosis remains guarded, and late recurrent ischemic events may be due to both restenosis of the target lesion and diffuse vein graft disease ^(382-384). Final patency after PTCA is greater for distal SVG lesions than for ostial or mid-SVG lesions ⁽³⁴⁹⁾, and stenosis location appears to be a better determinant of final patency than graft age or the type of interventional device used.

Percutaneous intervention for chronic vein graft occlusion has been problematic. Balloon angioplasty alone has been associated with high complication rates and low rates of sustained patency ⁽³⁸⁵⁾. Although prolonged intra-graft infusion of fibrinolytic therapy was reported to successfully recanalize 69% of a selected group of patients with chronic SVG occlusion less than 6 months duration, long-term patency rates with or without adjunctive stent deployment were low ^(386-388). In addition, prolonged fibrinolytic therapy has been associated with thromboembolic MI ^(389-392), intracranial ⁽³⁹³⁾ and intramyocardial hemorrhage ⁽³⁹⁴⁾, as well as vascular access site complications. Favorable results have been obtained with both local “targeted” and more prolonged infusion of fibrinolytic agents for nonocclusive intragraft thrombus ^(395,396). Thrombolytic catheter-based systems appear to successfully treat SVG thrombosis as well as or better than thrombolytic agents ⁽³⁹⁷⁾.

3. Early and Late Outcomes of Percutaneous Intervention. Prior to the general availability of coronary stenting, overall angioplasty procedural success rates exceeded 90%, and adverse outcomes of emergency repeat coronary bypass surgery (2.3%) and death (0.8%) were infrequent as reported in combined series of over 2000 patients with prior bypass surgery undergoing percutaneous intervention ^{(135,398-410)}. These results are comparable to those achieved in patients without prior bypass surgery, an observation confirmed by NHLBI registry data ⁽⁶⁾. The most common complications observed in this population are NSTEMI and atheroembolism, particularly following SVG intervention ^(333,352).

Patients with prior bypass surgery who undergo successful PCI have a long-term outcome that is dependent on patient age, the degree of LV dysfunction, and the presence of multivessel coronary atherosclerosis. The best long-term results are observed after recanalization of distal anastomotic stenoses occurring within 1 year of operation. Angioplasty of distal anastomotic stenoses involving internal mammary artery grafts have been associated with similar, favorable long-term patency rates ^(357,358). Conversely, event-free survival is less favorable following angioplasty of totally occluded SVGs, ostial vein graft stenoses, or grafts with diffuse or multicentric disease ^{(382,383,385)}. Coexistent multisystems disease, the presence of which may have prompted the choice of a percutaneous revascularization strategy, may also influence long-term outcomes in this population.

4. Surgery Versus Percutaneous Reintervention. Aged, diffuse, friable, and degenerative SVG disease in the absence of a patent arterial conduit to the left anterior descending artery represents a prime consideration for repeat surgical revascularization. In contrast, the presence of a patent arterial conduit to the left anterior descending artery may militate for a percutaneous interventional approach ⁽⁴¹¹⁾. The overall risk of repeat operation, especially the presence of comorbidities such as concomitant cerebrovascular, renal, or pulmonary disease and the potential for jeopardizing patent, nondiseased bypass conduits must be carefully considered. Isolated, friable stenoses in vein grafts may be approached with primary stenting or the combination of extraction atherectomy and stenting in an attempt to reduce the likelihood of distal embolization.

Another therapeutic option for patients with prior coronary bypass surgery that has become available is grafting using the internal mammary artery through a “minimally invasive” surgical approach ^{(158,412-416)}. This strategy, which avoids both the risk of cardiopulmonary bypass (stroke, coagulopathy) and repeat median sternotomy may be particularly applicable to patients with chronic native vessel left anterior descending coronary occlusion and friable atherosclerotic disease involving a prior SVG to this vessel. The role of combining a minimally invasive surgical approach with PCI requires further study ^(417,418).

In general, patients with multivessel disease, failure of multiple SVGs, and moderately impaired LV function derive the greatest benefit from the durability provided by surgical revascularization with arterial conduits. Regardless of repeat revascularization strategy, risk-factor modification with cessation of smoking ^(419,420) and lipid lowering therapy ^(421,422) should be implemented in patients with prior CABG surgery. An aggressive lipid-lowering strategy that targets a low-density lipoprotein level of less than 90 mg/dL can be effective in reducing recurrent ischemic events and the need for subsequent revascularization procedures ⁽⁴²²⁾.

Recommendations for PCI With Prior CABG (Table 28)

Class I
1. Patients with early ischemia (usually within 30 days) after CABG ⁽¹⁹⁴⁾. (Level of Evidence: B)

Class IIa
1. Patients with ischemia occurring 1 to 3 years postoperatively and preserved LV function with discrete lesions in graft conduits. (Level of Evidence: B)

2. Disabling angina secondary to new disease in a native coronary circulation. (If angina is not typical, the objective evidence of ischemia should be obtained.) (Level of Evidence: B)

3. Patients with diseased vein grafts >3 years following CABG. (Level of Evidence: B)

Class III
1. PCI to chronic total vein graft occlusions. (Level of Evidence: B)

2. Patients with multivessel disease, failure or multiple SVGs, and impaired LV function. (Level of Evidence: B)

E. Use of Adjunctive Technology (Intracoronary Ultrasound Imaging, Flow Velocity, and Pressure)

The limitations of coronary angiography for diagnostic and interventional procedures can be reduced by employing adjunctive technology of intracoronary ultrasound imaging, flow velocity, and pressure. Information obtained from the adjunctive modalities of intravascular imaging and physiology can improve PCI methods and outcomes.

1. Intravascular Ultrasound Imaging (IVUS). IVUS imaging provides a tomographic 360° sagittal scan of the vessel from the lumen through the media to the vessel wall. IVUS measurements of arterial dimensions (minimal and maximal diameters, cross-sectional area, and plaque area) complement and enhance angiographic information. IVUS has been used to refine device selection through plaque characterization (e.g., calcified) and artery sizing. IVUS has contributed to the understanding of mechanisms of coronary angioplasty and specifically, to the advancement of coronary stenting without long-term anticoagulation ^(423-428). In a large observational study, IVUS-guided angioplasty resulted in a decreased final residual plaque area from 51 to 34%, despite a final angiographic percent stenosis of 0% ⁽⁴²³⁾. IVUS-facilitated stent deployment was associated with a subacute thrombosis rate of 0.3% without systemic anticoagulation, although antiplatelet agents are still required for stenting ⁽⁴²³⁾. In the placement of coronary stents, because radiographic contrast material can be located between stent struts and the vascular wall, an angiographic appearance of a large lumen may exist when the stent has not been fully deployed. IVUS documents full apposition of stent struts to the vessel wall ⁽⁴²³⁾.

IVUS is not necessary for all stent procedures. The results of the French Stent Registry study of 2900 patients treated without coumadin and without IVUS reported a subacute closure rate of 1.8% ⁽⁴²⁹⁾. In the STARS trial ⁽³⁰⁾, a subacute closure rate of 0.6% in patients having optimal stent implantation supports the approach that IVUS does not appear to be required routinely in all stent implantations. However, the use of IVUS for evaluating results in high-risk procedures (i.e., those patients with multiple stents, impaired TIMI grade flow or coronary flow reserve, and marginal angiographic appearance) appears warranted.

The long-term outcomes when adjunctive IVUS is used are currently under study. In a trial of 161 patients (MUSIC Trial) ⁽⁴³⁰⁾ evaluating optimal stent expansion (defined as complete apposition of the stent over its length) with symmetrical expansion (defined as a luminal diameter_minimum to luminal diameter_maximum >0.7) and minimal luminal area (compared to >80% of the reference area), the subacute closure rate was 1.3% with monotherapy of aspirin. The angiographic restenosis was <10% when stent cross-sectional areas were >9.0 mm².

Fitzgerald et al. report that the degree of stent expansion as measured by IVUS directly correlates to the clinical outcomes in the CRUISE study ⁽⁴³¹⁾. This multicenter study compared 270 patients with IVUS-guided stent implantation to IVUS-documented, but not guided, stent implantation in 229 patients. At 9-month follow-up, there was no difference in death or myocardial infarction rate, but the target lesion revascularization rate was substantially lower in the IVUS-guided group (8.5% vs. 15.3%; p = 0.019). These data suggest that ultrasound guidance of stent implantation may result in more effective stent expansion compared with angiographic guidance alone and subsequently reduced the need for late target lesion revascularization.

In the context of published data and growing clinical experience, the Writing Committee has modified prior recommendations for the use of IVUS as follows.

Recommendations for Coronary Intravascular Ultrasound (Table 29)

Class IIa
1. Assessment of the adequacy of deployment of coronary stents, including the extent of stent apposition and determination of the minimum luminal diameter within the stent. (Level of Evidence: B)

2. Determination of the mechanism of stent restenosis (inadequate expansion vs. neointimal proliferation) and to enable selection of appropriate therapy (plaque ablation vs. repeat balloon expansion). (Level of Evidence: B)

3. Evaluation of coronary obstruction at a location difficult to image by angiography in a patient with a suspected flow-limiting stenosis. (Level of Evidence: C)

4. Assessment of a suboptimal angiographic result following PCI. (Level of Evidence: C)

5. Diagnosis and management of coronary disease following cardiac transplantation. (Level of Evidence: C)

6. Establish presence and distribution of coronary calcium in patients for whom adjunctive rotational atherectomy is contemplated. (Level of Evidence: C)

7. Determination of plaque location and circumferential distribution for guidance of directional coronary atherectomy. (Level of Evidence: B)

Class IIb
1. Determine extent of atherosclerosis in patients with characteristic anginal symptoms and a positive functional study with no focal stenoses or mild CAD on angiography. (Level of Evidence: C)

2. Preinterventional assessment of lesional characteristics and vessel dimensions as a means to select an optimal revascularization device. (Level of Evidence: C)

Class III
1. When angiographic diagnosis is clear and no interventional treatment is planned. (Level of Evidence: C)

2. Coronary Flow Velocity and Coronary Vasodilatory Reserve. Coronary physiologic information has assumed increasing importance in determining which coronary lesions may merit intervention and achieve an endpoint of balloon angioplasty in consideration of provisional stenting. Coronary flow velocity reserve (CVR), the ratio of hyperemic to basal flow, reflects flow resistance through the epicardial artery and the corresponding myocardial bed. CVR <2.0 is reproducibly and positively correlated to abnormal stress perfusion imaging ^(432-434). In some cases, the uncertainty as to whether the impaired flow reserve is due to the target stenosis or to abnormal microcirculation may be reduced using a relative coronary flow velocity reserve (rCVR; CVR_target/CVR_reference). From preliminary studies, rCVR values >0.8 are similar in prognostic value to negative stress testing ⁽⁴³⁵⁾. Recent confirmation of the more lesion-specific nature of physiologic indices resides in the correlation between rCVR and pressure-derived fractional flow reserve (FFR) of the myocardium by guidewire pressure measurements (vide infra) ^(435,436).

For lesion assessment, a normal CVR indicates a nonphysiologically significant stenosis. An abnormal CVR indicates that the stenosis in the epicardial artery is significant when the microcirculation is normal, confirmed by measuring rCVR. Several studies report that deferring PCI of non-flow-limiting lesions is safe, with <10% rate of lesion progression ^(437-439).

3. Coronary Artery Pressure and Fractional Flow Reserve. Historically, translesional pressure gradients were used as endpoints for early interventional cardiology procedures. The use of a translesional pressure gradient measured at rest was abandoned because of weak correlations to stress testing and difficult technique. Pijls et al. ⁽⁴⁴⁰⁾ introduced the concept of the fractional flow reserve (FFR) of the myocardium, the ratio of distal coronary pressure to aortic pressure measured during maximal hyperemia, which represents the fraction of normal blood flow through the stenotic artery ^(436,441). The coronary pressure measuring technique is relatively simple, especially using pressure guidewires, a method superior to small catheters. The normal FFR value for all vessels under all hemodynamic conditions, regardless of the status of microcirculation is 1.0. FFR values <0.75 are associated with abnormal stress tests ⁽⁴³⁸⁾. Unlike CVR, the FFR is relatively independent of hemodynamics and microcirculatory disturbances. FFR does not use measurements in a reference vessel and is thought to be epicardial lesion-specific. FFR provides no information on the microcirculation nor on the absolute magnitude of the change in coronary flow.

Reports indicate that a physiologic assessment can determine whether balloon angioplasty alone has achieved a satisfactory result with 6-month outcome equivalent to that reported with elective stenting. The DEBATE trial ⁽⁴⁴²⁾ in 224 patients found that when a final diameter stenosis <35% and an excellent physiologic result (CVR >2.5) were obtained after balloon angioplasty (44/224 patients), the intermediate-term (6 months) target lesion revascularization and angiographic restenosis rates were 16%. Similar data have been reported for FFR ⁽⁴³⁹⁾. The application of coronary physiologic adjunctive modalities can facilitate decision-making for moderate lesions, the appropriateness of balloon angioplasty, and the use of provisional stenting.

Recommendations for Intracoronary Physiologic Measurements (Doppler Ultrasound, FFR) (Table 30)

Class IIa
1. Assessment of the physiological effects of intermediate coronary stenoses (30 to 70% luminal narrowing) in patients with anginal symptoms. Coronary pressure or Doppler velocimetry may also be useful as an alternative to performing noninvasive functional testing (e.g., when the functional study is absent or ambiguous) to determine whether an intervention is warranted. (Level of Evidence: B)

Class IIb
1. Evaluation of the success of percutaneous coronary revascularization in restoring flow reserve and to predict the risk of restenosis. (Level of Evidence: C)

2. Evaluation of patients with anginal symptoms without an apparent angiographic culprit lesion. (Level of Evidence: C)

Class III
1. Routine assessment of the severity of angiographic disease in patients with a positive, unequivocal noninvasive functional study. (Level of Evidence: C)