Pharmacological Facilitation of Primary Percutaneous Coronary Intervention for Acute Myocardial Infarction: Title and subTitle BreakIs the Slope of the Curve the Shape of the Future?

In the current era, there is general consensus that primary percutaneous coronary intervention (PCI) is the preferred approach to reperfusion therapy when delivered expeditiously in centers with requisite facilities and documented expertise and outcomes.¹ Nonetheless, in patients presenting very early (within 2 hours) after symptom onset, rapid administration of fibrinolytic therapy is associated with excellent outcomes and may be preferred, particularly in centers without facilities for primary PCI² or when primary PCI is likely to be delayed, causing first balloon inflation to occur more than 60 minutes after initiation of thrombolytics.³ Moreover, the logistics of delivering primary PCI at all times are complex and variable among different regions and centers.⁴

Options for patients admitted to community hospitals include administration of fibrinolytics followed by watchful waiting, primary PCI without on-site cardiac surgery, transfer to a tertiary care center for primary PCI, or a strategy of facilitated PCI.⁵ Facilitated PCI involves administration of fibrinolytics and glycoprotein IIb/IIIa inhibitors (alone or in combination) in an attempt to restore partial infarct-related artery blood flow early (before PCI is possible), followed by early transfer for immediate (as opposed to elective) PCI for reperfusion in patients in whom pharmacological therapy has failed and for definitive revascularization.⁵ In this context, transfer for proposed PCI with adjunctive antithrombotic therapy, but in the absence of thrombolytics or glycoprotein IIb/IIIa inhibitors (or both), does not fall under the rubric of facilitated angioplasty. Crucial to these options is an understanding of the dynamic evolving relationship among reperfusion, myocardial salvage, and mortality reduction.⁶

The purpose of this review of the pathophysiology of reperfusion therapy is to derive insights about the likelihood of success of facilitated PCI compared with transfer-only or fibrinolysis-only strategies. The feasibility and safety of primary PCI without on-site cardiac surgery in patients with acute coronary syndromes has been demonstrated.^{7 - 9} Successful performance of PCI in centers without cardiac surgery is facilitated by the establishment of algorithms, training of personnel, meticulous documentation of outcomes, and regular audits, but not reliance on published results of other centers.

Theoretical Considerations of Time Dependence of Reperfusion Therapy. A fundamental tenet of reperfusion therapy is that time is of the essence because ischemia leading to necrosis is a progressive process.^{6 ,10 - 11} The curve describing this relationship shows a striking benefit within the first 2 to 3 hours, emphasizing the narrowness of the “golden window of opportunity,” followed by a continued mortality benefit of decreasing magnitude over time (Figure).⁶ Whereas the major mechanism of early benefit is likely a consequence of myocardial salvage, the extent to which the “open artery” exerts a beneficial impact independent of any effect on salvage remains controversial, awaiting results of ongoing trials.^{12 - 15} Moreover, the specific time period for reperfusion in humans during which salvage occurs is unknown and subject to modifying influences, including presence of functioning collaterals, myocardial oxygen demands, ischemic preconditioning, and the duration of sustained ischemia (Figure).

Time to Treatment. Despite recent advances, the goal of optimal reperfusion remains enticing but distant. The ability of fibrinolytic therapy to achieve TIMI (Thrombolysis in Myocardial Infarction) grade 3 flow and reduce mortality without increasing bleeding complications appears to have plateaued. Moreover, the risk-benefit ratio of adjunctive antithrombotic strategies is narrowing, and the achievement of overall “myocardial” perfusion is disappointing.

Previously, the metric of reperfusion therapy was the achievement of TIMI grade 3 flow. Multiple recent studies have documented, however, a discrepancy between the establishment of TIMI grade 3 flow and optimal myocardial perfusion.^{16 - 19} Several new grading schemes, including ST-segment resolution and angiographic myocardial blush, have been developed as surrogates of myocardial perfusion and have been shown to provide powerful incremental prognostic information.^{18 ,20 - 21} In addition, these surrogate markers appear to correlate strongly with time to treatment. The concept that “time is muscle” is well accepted but should be expanded to include the microvasculature and myocyte perfusion.

The superiority of primary PCI over thrombolytics alone may be reduced by transport delays.^{22 - 27} A meta-analysis of trials of transfer for primary angioplasty vs thrombolysis suggested that primary PCI is superior,²⁸ but this conclusion may be weighted heavily by the prolonged duration of symptoms before admission or randomization in many patients. Patient transfer times in the United States are far longer than in European trials such as the Danish Multicenter Randomized Study on Fibrinolytic Therapy vs Acute Coronary Angioplasty in Acute Myocardial Infarction (DANAMI) 2.²⁸ Obstacles to rapid transfer include greater distances between community and tertiary hospitals in the United States, lack of integrated medical services, inclement weather, and limited experience with centralized myocardial infarction networks.²⁹

A meta-analysis of trials of primary PCI vs thrombolytics suggested that the benefits of primary PCI are lost once delay between balloon inflation and administration of fibrinolytics exceeds 1 hour.³ Moreover, patients with some modicum of flow in the infarction-related artery before intervention have improved outcomes.^{30 - 32} This led to the logical strategy of “facilitated angioplasty” for expanding the benefits of primary PCI to community hospitals without on-site catheterization laboratories.

The impetus driving an approach of facilitated PCI is multifactorial. First, the correlation between symptom onset–to–balloon time and outcomes is strong. In contrast, the relation between door-to-balloon time and outcomes is controversial, but in patients considered at higher risk, mortality has been shown to increase substantially among those treated more than 4 hours after symptom onset^{25 - 26} and more than 3 hours afterward in another trial.²⁴ Another study demonstrated that after multivariable adjustment, each 30 minutes of delay was associated with a relative risk of 1-year mortality of 1.075 (95% confidence interval [CI], 1.008-1.15; P = .04).³³ However, the National Registry of Myocardial Infarction (NRMI) demonstrated discordance between symptom onset–to–balloon time, which did not correlate with mortality, in contrast with the strong relationship between door-to-balloon time and mortality (especially with >2-hour delays).²² In large multicenter studies, door-to-balloon time possibly is a surrogate for quality of care, and the time between symptom onset and start of therapy is the major determinant of the duration of ischemia and extent of salvage. An alternative explanation is that door-to-balloon times are dependent on the severity of illness, in that “sicker” patients may have longer door-to-balloon times because of the need for additional procedures and a prolonged time to reperfusion.

Logically, the impact of shorter door-to-balloon times is likely to be greatest in patients presenting early after onset of symptoms (<1-2 hours), as was the case in the Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications (CADILLAC) trial.²⁴

Impact of Preprocedure TIMI Flow. Restoration of infarction-related artery blood flow, particularly TIMI grade 3 flow, is associated with improved outcome whether reperfusion is spontaneous or follows pretreatment with fibrinolytics.³⁴

Previous Studies of Immediate Angioplasty After Thrombolytic Therapy. Various strategies of facilitated PCI are the subject of ongoing trials. Between 1987 and 1997, trials comparing angiography, with routine angioplasty performed early after thrombolytic therapy, and a conservative, ischemia-driven approach demonstrated variable differences in death and myocardial infarction and ejection fraction (Table 1).^{35 - 38} Recurrent ischemic events, however, were significantly reduced by the use of an aggressive approach. However, these trials occurred in an era reflecting a mechanical and pharmacological “learning curve” of PCI.⁴² These trials antedated the use of steerable wires, low-profile guiding catheters, stents, distal protection and thrombectomy devices, low osmolar contrast agents, and high-resolution imaging. Moreover, the adjunctive pharmacological therapy used in these earlier trials is obsolete in that platelet inhibitors, thienopyridines, direct thrombin inhibitors, and liberal use of statins had not been adopted; indeed, some trials did not routinely administer aspirin or its use was delayed for 24 hours.³⁵ Other advances, including use of smaller sheaths, lower doses of heparin, vascular closure devices, and point-of-care monitoring of activated clotting times will likely reduce bleeding. Two recent nonrandomized analyses of early angiography after fibrinolytic therapy showed safety and a trend toward efficacy and suggest that contemporary outcomes are substantially better.^{43 - 44} The results of several recently presented small randomized trials are described below and in Table 1 and Table 2.

Pilot Studies of Facilitated Angioplasty. Several pilot studies of facilitated PCI used reduced-dose thrombolytic agents (Table 2). In the Plasminogen-Activator Angioplasty Compatibility Trial (PACT), patients were assigned to reduced-dose tissue plasminogen activator (50 mg) vs placebo in addition to aspirin and heparin, followed by early angiography.³⁴ Patients in the tissue plasminogen activator arm had a higher rate of TIMI grade 3 flow at angiography (33% vs 18%; P<.001), and the discharge ejection fraction was highest among patients with preintervention TIMI grade 3 flow in contrast with those in whom this was achieved only after intervention. Clinical outcomes were similar in the 2 groups on an intention-to-treat basis.

A trend toward less favorable outcomes with facilitated PCI compared with primary percutaneous transluminal coronary angiography was noted in 2 trials using intravenous streptokinase^{47 - 48} and in 1 arm of a small trial using alteplase.⁴⁹ Combination therapy with a fibrinolytic and glycoprotein IIb/IIIa inhibitor produced a modest increase in TIMI grade 3 flow rates in 2 small trials.⁵⁰ In the Bavarian Reperfusion Alternatives Evaluation (BRAVE), 253 patients were assigned before referral for primary PCI to half-dose reteplase plus abciximab vs abciximab alone.⁴⁵ Initial TIMI grade 3 flow rates were higher with combination therapy (40% vs 18%; P<.001), but there was no difference in final infarction size, and bleeding complications were significantly increased in the combination group. Lack of reduction in infarction size may reflect that although drugs were administered approximately 2.5 hours after symptom onset, flow may not be restored until 30 to 45 minutes after drug administration, a time at which opening the artery is more important than the time it takes to achieve reperfusion (Figure).

Another recently reported trial compared 2 approaches; namely, primary PCI alone vs tenecteplase and enoxaparin followed by PCI within 3 to 12 hours (facilitated PCI). The latter approach appeared feasible, safe, and superior to PCI alone regarding complete ST-segment resolution at 6 hours (43% vs 61%, respectively; P = .03), although clinical outcomes were unchanged.⁵¹

Several trials evaluated glycoprotein IIb/IIIa inhibitors before PCI and demonstrated a modest benefit at best.^{25 ,52 - 54} In the large, multicenter Ongoing Tirofiban in Myocardial Infarction Evaluation (ON-TIME) trial, administering tirofiban before transfer for primary PCI had no clinical advantage.⁴⁶

Potential Outcomes of Facilitated PCI. The choice of pharmacological strategies for the facilitation of PCI is complex, with several logical candidates, including fibrinolytics in either full or reduced dose, glycoprotein IIb/IIIa inhibitors alone or in combination, unfractionated heparin, low-molecular-weight heparins, and direct thrombin inhibitors. The combination of fibrinolytics with platelet glycoprotein IIb/IIIa inhibitors has raised concerns about bleeding risks of subsequent “rescue” angioplasty, but there is a paucity of data. In this setting, aspirin and intravenous β-blockers would be considered standard-of-care adjunctive therapy.⁵⁵ Irrespective of the method of facilitation, it should be stressed that the concept of facilitated PCI, while logical, is unproven. Critical to the results of facilitated PCI is the slope of the curve in the Figure; specifically, the region of the curve occupied by the patient population being studied. Using this hypothetical construct, 3 different scenarios can be envisioned.

Patient Presentation Early After Onset of Symptoms. With patients presenting early (within 60-90 minutes) after onset of symptoms, the primary objective is to initiate reperfusion therapy as urgently as possible with minimal delay. In the National Registry of Myocardial Infarction (NRMI), approximately 60% of patients treated with reperfusion therapy presented within 2 hours after symptom onset.⁵⁶ The data from the Assessment of the Safety and Efficacy of a New Thrombolytic Regimen (ASSENT) 3 trial are less salutary in that the time from symptom onset to treatment was less than 2 hours for only 27% of patients and was within 1 hour for 3.2%.⁵⁷

Whether the therapy given is primary PCI or fibrinolytics depends on the policy of an individual institution, but speed is of the essence. In this respect, the results of the Comparison of Angioplasty and Prehospital Thrombolysis in Acute Myocardial Infarction (CAPTIM) trial of prehospital thrombolytic therapy (plus rescue PCI in approximately 26% of patients) vs primary PCI are relevant. In patients treated within 2 hours, mortality (P = .06) and development of cardiogenic shock were reduced by prehospital thrombolytics, but thereafter, the mortality trend favored primary PCI (P = .47).⁵⁸ The Myocardial Infarction Triage Intervention (MITI) trial also demonstrated the striking benefits of very early thrombolytic reperfusion therapy within 70 minutes after symptom onset.² Mortality in patients treated early vs later was 1.2% vs 8.7%, infarction size was 4.9% vs 11.2% of the left ventricle, and the predischarge ejection fraction was 53% vs 40%, respectively. Of patients treated with fibrinolytics in another trial, 13.3% had an “aborted” myocardial infarction; this was a remarkable 25% for patients treated within 1 hour of symptom onset.⁵⁷

This observation is consistent with the 50% mortality reduction in the Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto Miocardico (GISSI) 1 trial.^{11 ,59} Moreover, in the Primary Angioplasty in Acute Myocardial Infarction Patients From General Community Hospitals Transported for Percutaneous Transluminal Coronary Angioplasty Units vs Emergency Thrombolysis (PRAGUE) 2 trial, transfer for PCI was superior to fibrinolytic therapy only in the group treated 3 hours after symptom onset.⁶⁰ Nonetheless, these data must be interpreted cautiously because there is a paucity of comparative data within the first 90 to 120 minutes.

Patient Presentation After 2 Hours. In the Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) 1 and 3 trials, the average time from symptom onset to hospital arrival was 84 minutes. This did not change between 1990 and 1997,⁶¹ during which door-to-needle time shortened from 66 to 48 minutes; thus, symptom-to-treatment time ranged from 132 to 150 minutes. In the NRMI 3 study in 1999, the median door-to-needle time was 34 minutes.⁶² This translated into a delay of approximately 2 hours between symptom onset and treatment. In the NRMI 2 study, the mean time from symptom onset to hospital arrival was 114 to 126 minutes (median, 90-102 minutes).⁶³ Thus, many patients may present at points A and B (Figure), when time dependence of the extent of myocardial salvage is less important. In patients presenting more than 2 hours after symptom onset, Schomig et al⁶⁴ demonstrated the superiority of primary PCI over tissue plasminogen activator therapy for myocardial salvage and infarction size, a finding explicable on the basis of thrombus resistance to fibrinolytics over time, in contrast with sustained efficacy of the mechanical approach.⁶⁵

Moreover, the benefits of thrombolytics late after symptom onset are decreased and the risk of myocardial rupture may be increased.^{66 - 67} In other words, at points A and B on the curve, the emphasis is on restoring arterial patency even though time to treatment is longer. This concept is supported by data from the CADILLAC trial of patients undergoing primary PCI. Survival was greatest when reperfusion was obtained within 3 hours after symptom onset, whereas incremental delays thereafter had little additional mortality effect.²⁴ Because thrombolytic therapy requires, on average, 45 to 60 minutes before reperfusion occurs,⁶⁸ a case can be made for transfer for PCI without preceding fibrinolytic therapy in patients presenting relatively late (ie, ≥3 hours after symptom onset).²⁸

The crucial question is whether facilitation affects the results of primary PCI in patients presenting 2 to 3 hours after symptom onset. If facilitation moves patients from the plateau to the descending limb of the curve (eg, a shift from point B to point C and from point A to point C in the Figure), a benefit can be expected. The magnitude of benefit and its clinical relevance depends on how far up the curve the population can be shifted. Thus, a shift from point A to point B is unlikely to manifest as a change in any clinically relevant end point. This may explain the results of the BRAVE trial, because the median time from symptom onset to drug administration was 160 to 164 minutes.⁴⁵ Modifying factors such as myocardial oxygen demands, ischemic preconditioning, collateral circulation, and the dynamics of thrombosis and intermittent lysis may be important.

Facilitated PCI may improve reperfusion therapy, but not without risk, and there are pitfalls. Whether benefit outweighs risk depends largely on the duration of symptoms or ischemia and the nature of adjunctive therapy (ie, full- or reduced-dose fibrinolytics, platelet inhibitors, or heparin alone). The risk of intracranial bleeding must be considered, particularly among elderly persons, women, patients with hypertension, and those with low body weight.⁶⁹ The sequelae of access-site hemorrhage following femoral angiography shortly after fibrinolytic therapy are not minor issues. Moreover, early reperfusion (TIMI grade 3 flow) is unlikely to occur in more than 60% of patients with thrombolysis before PCI, compared with approximately 10% to 20% of patients who achieve TIMI grade 3 flow before angioplasty from endogenous thrombolysis.

Conversely, transfer for PCI without preceding fibrinolytic therapy could be harmful if the delays resulted in a shift in the curve from point D to point C or point B (Figure)—a situation applicable to patients presenting within 1 to 2 hours after symptom onset. For patients presenting more than 2 to 3 hours after symptom onset, in whom the likelihood of facilitation with fibrinolytics resulting in beneficial outcomes is slim, administration of glycoprotein IIb/IIIa inhibitors before transfer is reasonable because the risk of intracranial hemorrhage is small and the early safety profile of primary PCI is enhanced.⁷⁰ However, more data are needed. Crucial components of the clinical evaluation include an assessment of the risk of bleeding with fibrinolytic agents, the hemodynamic stability of the patient, and a realistic estimation of the delay incurred by transfer.

Alternative Approach to Facilitated PCI. An alternative approach to facilitated PCI is to treat all patients in community hospitals without facilities for primary PCI with a fibrinolytic or prehospital fibrinolytic strategy, followed by transfer to a referral hospital. A case may be made for direct transfer without prior administration of fibrinolytics in patients presenting more than 2 or 3 hours after symptom onset, as discussed previously. Moreover, in elderly persons, the risk of intracranial hemorrhage is a major factor underlying the risk-benefit ratio of fibrinolytics vs primary PCI.

Whether age 75 to 80 years or older might itself be an indication for transfer without preceding fibrinolytic therapy is a reasonable question for which definitive answers are lacking.⁷¹ For cardiogenic shock, which is characteristically a dynamic and evolving process, the prognosis is so ominous that it is reasonable to attempt some degree of reperfusion using fibrinolytics before transfer.⁷² Other factors warranting consideration are whether myocardial infarction is anterior or inferior and the presence of hemodynamic and electrical stability.⁷³

Following fibrinolytic therapy, the next step after transfer would be “rescue” PCI in selected patients, as demonstrated in the CAPTIM trial.⁵⁸ Transfer for rescue PCI should not be delayed until treatment failure is obvious and the patient is hemodynamically compromised. A high index of suspicion for fibrinolytic failure needs to be maintained, and, when in doubt, transfer should occur earlier. Percutaneous coronary intervention is logical, but proof of the concept has long been elusive.^{74 - 75} For the majority of stable patients, whether subsequent management should be invasive or conservative is controversial. The Combined Angioplasty and Pharmacological Intervention vs Thrombolysis Alone in Acute Myocardial Infarction (CAPITAL AMI) trial addressed one version of such an approach and suggested that early angiography following fibrinolytics was safe and effective.³⁹ Similar results were obtained in a German trial that compared early vs delayed angiography after fibrinolytics.⁴⁰ The Grupo de Analisis de la Cardiopatia Isquemica Aguda (GRACIA) 1 trial documented the feasibility and safety of routine angiography within 24 hours after symptoms, which resulted in a significant reduction in the duration of hospitalization and in the primary combined end point of death, reinfarction, and revascularization at 12 months.^{41 ,76} Given the frequency and negative impact of reinfarction in addition to data that suggest that this is more likely to occur in patients with severe grades of residual stenosis,⁷⁷ predischarge angiography is appropriate; however, the cost implications of the procedure itself and the need for interhospital transfer justify additional randomized trials. If transfer is contemplated, logically, it should be to a hospital with interventional facilities, although such an approach appears to be highly variable within the United States.⁷⁸

These different approaches to reperfusion therapies need to be placed within the context of the debate on the regionalization of acute myocardial infarction care, both in the United States and in other countries. The preferred strategy, whether to transfer to an interventional facility or to the nearest hospital in a particular region, depends heavily on logistical constraints, including transfer times.

Two key concepts discussed in this review are of profound importance: (1) the Reimer and Jennings wavefront hypothesis of evolving infarction, which has driven the entire field of reperfusion therapy for almost 3 decades¹⁰ ; and (2) the curve of the relationship between time to treatment and extent of myocardial salvage in the clinical arena. These issues should give pause for thought as new strategies are developed. It is the intersection of these 2 concepts that gives rise to the current controversies surrounding reperfusion and forms the basis for the question, “Is the slope of the curve the shape of the future?”

Facilitated PCI is a logical concept that requires rigorous scrutiny of randomized controlled trials encompassing a wide spectrum of patients who present within 2 to 3 hours after symptom onset. Hopefully, the 2 large ongoing trials, Facilitated Intervention With Enhanced Reperfusion Speed to Stop Events (FINESSE)⁷⁹ and Assessment of the Safety and Efficacy of a New Thrombolytic Regimen (ASSENT) 4 (Allan Ross, MD, oral communication, June 2004), will result in another step up on the ladder of progress in reperfusion therapy. Nonetheless, to place this potential advance in perspective, the results of recent large registry studies are sobering and disconcerting.⁸⁰ Approximately 25% to 30% of eligible patients in the Global Registry of Acute Coronary Events (GRACE) do not receive any form of reperfusion therapy; time to presentation appears to have plateaued, and door-to-treatment times remain disappointingly prolonged.

The issue of facilitated PCI is important and worthy of effort and further study, and the cardiology community eagerly awaits the results in the justifiable expectation that we may have definitive answers to important new questions. The clinical impact of these trials will be enhanced if the patient population enrolled encompasses a wide spectrum of risk. What is unknown is whether novel therapies will be discovered with the ability to shift the curve to enhance the degree of myocardial salvage at any given time of reperfusion.

Nonetheless, the major goal from a societal perspective remains as true today as at the start of the reperfusion era and will not be changed by the results of trials of facilitated PCI. This goal is to reperfuse an occluded coronary artery in as many eligible patients as possible, as rapidly and safely as possible. The focus must not be only on the nature of the therapy but also on the efficacy and safety of its delivery.