Author + information
- Received January 13, 2000
- Revision received June 6, 2000
- Accepted June 21, 2000
- Published online November 1, 2000.
- Howard C Herrmann, MD∗,*,
- David J Moliterno, MD†,
- E.Magnus Ohman, MD‡,
- Amanda L Stebbins, MS‡,
- Christopher Bode, MD§,
- Amadeo Betriu, MD∥,
- Florian Forycki, MD¶,
- Jerry S Miklin, MD#,
- William B Bachinsky, MD∗∗,
- A.Michael Lincoff, MD†,
- Robert M Califf, MD‡ and
- Eric J Topol, MD†
- ↵*Reprint requests and correspondence: Dr. Howard C. Herrmann, Hospital of the University of Pennsylvania, 3400 Spruce Street, 9 Founders Pavilion, Philadelphia, Pennsylvania 19104
We examined the utility of early percutaneous coronary intervention (PCI) in a trial that encouraged its use after thrombolysis and glycoprotein IIb/IIIa inhibition for acute myocardial infarction (MI).
Early PCI has shown no benefit when performed early after thrombolysis alone.
We studied 323 patients (61%) who underwent PCI with planned initial angiography, at a median 63 min after reperfusion therapy began. A blinded core laboratory reviewed cineangiograms. Ischemic events, bleeding, angiographic results, and clinical outcomes were compared between early PCI and no-PCI patients (n = 162), between patients with Thrombolysis in Myocardial Infarction (TIMI) flow grade 0 or 1 before PCI versus flow grade 2 or 3, and among three treatment regimens.
Early PCI patients showed a procedural success (<50% residual stenosis and TIMI flow grade 3) rate of 88% and a 30-day composite incidence of death, reinfarction, or urgent revascularization of 5.6%. These patients had fewer ischemic events and bleeding complications (15%) than did patients not undergoing early PCI (30%, p = 0.001). Early PCI was used more often in patients with initial TIMI flow grade 0 or 1 versus flow grade 2 or 3 (83% vs. 60%, p < 0.0001). Patients receiving abciximab with reduced-dose reteplase (5 U double bolus) showed an 86% incidence of TIMI grade 3 flow at ∼90 min and a trend toward improved outcomes.
In this analysis, early PCI facilitated by a combination of abciximab and reduced-dose reteplase was safe and effective. This approach has several advantages for acute MI patients, which should be confirmed in a dedicated, randomized trial.
Percutaneous coronary intervention (PCI) often is performed in patients with acute myocardial infarction (MI), but controversy exists about the optimal indications and timing (1). In patients first treated with thrombolysis, immediate angioplasty has not been shown to provide any
advantage over performing the procedure 18 to 48 h later (2) or 7 to 10 days later (3) or compared with a conservative strategy requiring noninvasive testing before angioplasty (4,5). Angioplasty techniques have advanced substantially since these trials, however, and they include the current widespread use of stents and potent antiplatelet therapies.
The recent Strategies for Patency Enhancement in the Emergency Department (SPEED, or the Global Use of Strategies To Open Occluded coronary arteries [GUSTO-4] Pilot) trial, a randomized study of abciximab with or without low-dose fibrinolytic therapy for acute MI, showed improved patency of the infarct-related artery at 60 to 90 min, compared with a standard lytic regimen (6). Because all patients in this trial had early angiography and PCI was encouraged, there is an opportunity to re-examine the utility of early PCI after thrombolysis in a contemporary setting. We used the term facilitated PCI to describe early, planned PCI after a pharmacological regimen intended to open the infarct-related artery. Our hypothesis was that facilitated PCI incorporating stents and platelet glycoprotein IIb/IIIa therapy would be safe and would improve the angiographic and procedural outcomes of patients with acute MI.
The trial enrolled 528 patients with ST-segment elevation acute MI within 6 h of chest-pain onset (6). Patients with pulmonary edema or shock, previous bypass surgery or stroke, severe hypertension, or contraindication to thrombolysis were excluded. In the initial dose-ranging phase of the trial, 304 patients were randomized in a 4:1 ratio to receive abciximab (0.25-mg/kg bolus and 12-h infusion at 0.125 μg/kg/min) with reteplase (5 U, 7.5 U, 10 U, 5 + 2.5 U, or 5 + 5 U boluses) or no reteplase. The two reteplase boluses were given 30 min apart. All patients received aspirin and intravenous heparin as a weight-adjusted bolus (60 U/kg) and additional boluses to achieve an activated clotting time ≥200 s.
In the second, dose-confirmation phase of the trial, 223 patients were randomized to receive full-dose reteplase (10 + 10 U) or the Phase 1 regimen that resulted in the highest rate of Thrombolysis In Myocardial Infarction (TIMI) grade 3 flow before PCI—abciximab with reduced-dose reteplase (5 + 5 U). In this phase, heparin was administered as 40 U/kg (maximum 4,000 U) in the combination arm and 70 U/kg in full-dose reteplase patients. The primary end point in both phases was the incidence of TIMI grade 3 flow in the infarct-related artery (7) at angiography 60 min to 90 min after treatment began, as assessed by the core laboratory.
Cardiac catheterization and intervention
Coronary angiography was performed as soon as possible after reperfusion therapy began; then PCI was performed at the investigator’s discretion using any approved techniques and devices. “Early” intervention was defined as PCI performed at the time of first angiography. Based on the median angiography time of 63 min, balloon inflation was estimated to have occurred by 90 min. All cineangiograms were reviewed in a blinded manner by the angiographic core laboratory (Cleveland Clinic Foundation). Angiographic variables collected before and after PCI included percent diameter stenosis, TIMI flow grade, corrected TIMI frame count (CTFC) (8), and the presence of distal embolization, no-reflow, dissection, or perforation.
Procedural success was defined as post-PCI residual stenosis ≤50% with TIMI grade 3 flow. Major outcomes included death, reinfarction, or urgent, repeat PCI or bypass surgery for ischemia. Bleeding complications within 14 days were categorized as major (any intracranial, retroperitoneal, or intraocular bleeding, or a decrease in hemoglobin >5 g/dl), intracranial hemorrhage, or requiring transfusion. A composite “clinical success” variable was defined as freedom from death, reinfarction, urgent revascularization, major bleeding, or transfusion at 30 days.
Categorical variables were summarized as frequencies with percentages, and continuous variables as means ± SD or medians with interquartile ranges. We compared patients who did and did not undergo early intervention by use of the Fisher exact test for discrete variables and the Wilcoxon rank-sum test for continuous variables. Within the cohort of patients receiving early PCI, those with TIMI flow grade 0 or 1 before PCI were compared with those with TIMI flow grade 2 or 3 before PCI. Finally, we used log likelihood chi-square and the Kruskal-Wallis test to compare early PCI data from the dose-finding and dose-confirmation phases for three treatment regimens: abciximab alone, full-dose reteplase alone, and abciximab with reteplase 5 + 5 U.
Angiography was performed within the intended 60-to-90-min period in 466 of the 528 patients (median, 63 min; interquartile range, 60 min to 70 min). Investigators reported that PCI was performed with first angiography in 323 patients (61% of the intention-to-treat cohort), which made up the facilitated-PCI study population. Of the 205 patients who did not have early PCI, 43 had PCI with an initial angiogram performed either before or after the qualifying 60-to-90-min period or were missing TIMI grade flow and were excluded from analysis. The remaining 162 patients formed the “no early PCI” comparison group.
Overall results of early PCI
In the early-PCI group, the site-reported procedural success rate was 88%, and 78% received intracoronary stents. The mean age of patients was 60 ± 12 years, and 75% were men (Table 1). Diabetes and hypertension were present in 18% and 33%, respectively. The right coronary was the most prevalent infarct-related artery. Major outcomes occurred in 5.6% of patients, which included death (3.4%), reinfarction (1.2%), and urgent revascularization for severe ischemia (1.6%). Within 14 days, 6.5% of this group had major bleeding, and 9.0% required blood transfusion. Other procedure-related complications are shown in Table 2.
The 162 patients who did not undergo early PCI did not differ significantly in clinical characteristics or infarct location from the early-PCI group, but their rates of reinfarction (4.9%), urgent revascularization (9.3%), the composite of major outcomes (16.0%), and transfusion (16.0%) were substantially higher (Table 1; Fig. 1). ⇓ The clinical success rate was 85.4% for early-PCI patients and 70.4% for those not undergoing early PCI (p < 0.001) (Fig. 2).
The TIMI flow grade before and after PCI was assessed by the core laboratory in 294 patients with evaluable angiograms. Before PCI, 99 patients (33.7%) had TIMI flow grade 0 or 1, and 195 patients (66.3%) had TIMI flow grade 2 or 3. The no-early-PCI patients were more likely to have patent arteries on the 60-to-90-min angiogram (TIMI flow grade 2 or 3 = 91%) than early-PCI patients (p < 0.001). After early PCI, the percentage of patients with TIMI flow grade 2 or 3 improved to 97.6% (p < 0.001) (Fig. 3).
Results by initial TIMI flow grade
Despite the absence of differences in baseline characteristics other than the difference in TIMI flow, patients with TIMI flow grade 0 or 1 before PCI (n = 99) underwent early PCI more often (83% vs. 60%, p < 0.0001), but they were less likely to have TIMI grade 3 flow or a normal corrected TIMI frame count after PCI (Table 2). There were no significant differences in 30-day clinical outcomes or bleeding complications.
Comparison by treatment assignment
By intention-to-treat, 63 patients were randomized to abciximab alone, 109 patients to full-dose reteplase alone (two 10 U boluses, 30 min apart), and 191 patients to abciximab and reteplase (two 5 U boluses, 30 min apart). Of these, 294 patients (81%) had evaluable angiograms, and 195 (66%) underwent early PCI.
Early PCI was performed more often in patients receiving abciximab alone, probably reflecting the lower initial rate of TIMI grade 3 flow. Stents were used most often in the full-dose reteplase group. After PCI, TIMI grade 3 flow was present in 84% to 95% of patients. In patients receiving combination therapy, which resulted in the highest rate of TIMI grade 3 flow (47%) at 60 to 65 min, post-PCI TIMI grade 3 flow was present in 86% at an estimated 90 min after treatment began.
Corrected TIMI frame count analysis confirmed the TIMI flow-grade data, with the lowest counts before PCI in patients receiving combination therapy. After early PCI, the mean CTFC was normal (<30) in all three treatment groups.
There was a trend toward improved 30-day outcomes in the patients receiving combination therapy, with a rate of death, MI, and urgent revascularization of 5.9%, compared with 8.1% and 7.1% in patients receiving abciximab or reteplase alone, respectively. Intracranial hemorrhage was rare in all three groups; however, a trend was apparent toward greater bleeding complications primarily at vascular access sites with combination therapy. Clinical success at 30 days was similar (85% to 89%) for all treatment regimens (Table 3).
The SPEED (GUSTO-4 Pilot) trial showed that the combination strategy of potent platelet inhibition (abciximab) and reduced-dose thrombolysis (reteplase) improves infarct-related artery patency at 60 to 90 min, compared with abciximab alone or full-dose reteplase (6). Unlike the TIMI-14 trial (9), which reached a similar conclusion using alteplase, intervention at the time of initial angiography was encouraged in SPEED, resulting in early PCI in 61% of enrolled patients.
Results of early PCI
The results of PCI in this setting appear to be both safe and effective. Procedural success was achieved in 88% of patients, with a 30-day composite rate of death, reinfarction, and urgent revascularization of 5.6%. Overall clinical success at 30 days—freedom from death, reinfarction, urgent revascularization, major bleeding, and transfusion—was achieved in 85% of patients undergoing early PCI. These results contrast with previous studies of angioplasty performed shortly after full-dose thrombolysis (2–5). Although these earlier trials varied in the thrombolytic agent used, the timing of intervention, and the outcome measured, their results were surprisingly similar. In all four trials, immediate angioplasty had a lower success rate, higher mortality, and higher rates of reinfarction, bypass surgery, and transfusion than did delayed intervention or a conservative approach (2–5).
The most likely reasons for the better results in the present trial include the use of abciximab, intracoronary stents, optimization of heparin anticoagulation, and increased operator experience. Although one recent trial utilizing a reduced dose lytic with early planned PCI demonstrated no increase in the acute frequency of adverse events—despite a low usage of stents (26%) and abciximab (5%) (10)—platelet activation is an important contributing factor in thrombus formation after thrombolysis for acute MI (11). In this regard, several trials have shown that abciximab can improve the results of primary angioplasty in acute MI. In the ReoPro and Primary PTCA Organization and Randomized Trial (RAPPORT), 483 patients with acute MI were randomized to abciximab or placebo before angioplasty (12). The rates of death or MI and death, MI, or urgent revascularization were reduced from 11.2% to 8.7% and from 17.8% to 11.8%, respectively (12). A similar 47% reduction in a combined end point was observed in the ADMIRAL (Abciximab before Direct angioplasty and stenting in acute Myocardial Infarction Regarding Acute and Long-term follow-up) trial, which also allowed stent implantation (13). The benefit of stenting in acute MI also has been examined without abciximab and shown to be safe and to improve short-term outcomes (14,15). Other studies are under way that will address various combinations of abciximab, primary angioplasty, and primary stenting (16).
The specific fibrinolytic agent used, reteplase in this case, also may have contributed to the improved outcome. Reteplase has been shown to produce greater early TIMI grade 3 flow compared with alteplase (17). Although the GUSTO-III trial showed no difference in mortality rates of patients with MI treated with reteplase versus alteplase (18), patients undergoing rescue angioplasty with abciximab had improved outcomes when they had been randomized to reteplase compared with alteplase (19).
Benefits of facilitated PCI
Our data suggest multiple benefits from the use of combination therapy with abciximab and reduced-dose thrombolysis for “facilitated early PCI,” a term we propose to refer to planned PCI after pharmacological reperfusion therapy. Facilitated early PCI must be differentiated from direct or primary PCI (without thrombolysis); from immediate, early, or delayed PCI after full-dose thrombolytic administration; from rescue PCI after thrombolysis, with or without platelet glycoprotein IIb/IIIa inhibition; and from primary PCI with a glycoprotein IIb/IIIa inhibitor alone, either before or during PCI.
Early PCI was performed less often in patients receiving combination therapy (65% vs. 82% in patients receiving abciximab alone), which probably reflects the improved flow achieved before PCI in these patients. Early PCI was also performed less often in patients with improved flow (Table 2), and combination-therapy patients were more likely to have a patent artery at initial angiography. The possibility that an improved combination pharmacological regimen could lessen the need for early PCI suggests that this strategy might be cost-effective. In this regard, patients undergoing early PCI also had less major bleeding and required fewer transfusions by 30 days, probably reflecting the less frequent need for urgent revascularization including coronary artery bypass graft surgery.
Another advantage of facilitated early PCI relates to patient stability in the catheterization laboratory. Patients arriving in the laboratory with patent infarct-related arteries due to earlier reperfusion are less likely to be in cardiogenic shock, to require an intra-aortic balloon pump or temporary pacemaker, or to suffer cardiac arrest (20). Furthermore, a less hectic procedure and the ability to visualize the distal vessel may improve the technical success of angioplasty.
Finally, facilitated early PCI has the potential to fuse the best aspects of thrombolysis and primary angioplasty in the management of acute MI. A long-standing controversy has focused on which strategy provides the best outcome based on the incidence of TIMI grade 3 flow at 90 min (21–23). The time at which normal flow is achieved is equally important, however (24). A combined pharmacological approach using abciximab and reduced-dose reteplase or alteplase can improve TIMI grade 3 flow rates at 60 min by an absolute 10% to 15% (6–9). Primary angioplasty can achieve higher TIMI grade 3 flow rates than thrombolysis, but generally at later time points (23–25). The addition of early facilitated PCI in this trial between 60 and 90 min resulted in an extremely high core-laboratory-assessed TIMI grade 3 flow rate (85%) and a normal mean corrected TIMI frame count, without sacrificing the early benefit (between 30 and 60 min) of a pharmacological approach. We believe that the additive benefits of these two approaches could provide the optimal reperfusion strategy for many patients with acute MI (Fig. 4).
Early PCI was not mandated in this study and was therefore applied at each investigator’s discretion. This limits the comparison of early PCI patients with those not receiving PCI, as well as between different treatments, when the rate of early PCI varied. It is likely that physician selection influenced the use of early PCI, and it is possible that unmeasured, confounding variables contributed to the different outcomes observed in these comparisons. However, this is the largest study of this novel approach in a contemporary setting using stents and core-laboratory analysis. Our study used TIMI grade 3 flow as a primary end point. Although the Global Utilization of Streptokinase and TPA (alteplase) for Occluded Coronary Arteries (GUSTO-I) trial (25) confirmed a direct relation between TIMI grade 3 flow and survival, it is unknown whether further improvements in TIMI flow will result in greater survival (26). The recently reported discrepancy between myocardial tissue perfusion and epicardial coronary flow highlights this potential limitation of TIMI flow grade as a surrogate end point (27).
This study shows the safety and efficacy of early PCI in acute MI, facilitated by a combination of abciximab and reduced-dose reteplase. This approach has the potential to improve both very early and later reperfusion rates and to become the optimal reperfusion strategy. However, the mortality benefit and safety of combined therapy with abciximab and a low-dose fibrinolytic must first be validated in a large trial, such as the current GUSTO-4 acute MI trial, and the benefits of facilitated PCI should be confirmed in a dedicated, randomized trial.
☆ The SPEED (GUSTO-4 Pilot) trial was funded by grants from Eli Lilly and Company, Indianapolis, Indiana, and Centocor, Malvern, Pennsylvania.
- corrected TIMI frame count
- Global Utilization of Streptokinase and TPA (alteplase) for Occluded Coronary Arteries
- GUSTO-III, −4
- Global Use of Strategies To Open Occluded coronary arteries
- myocardial infarction
- percutaneous coronary intervention
- Strategies for Patency Enhancement in the Emergency Department
- Thrombolysis In Myocardial Infarction
- Received January 13, 2000.
- Revision received June 6, 2000.
- Accepted June 21, 2000.
- American College of Cardiology
- Vogt A,
- Neuhaus K.L
- SWIFT (Should We Intervene Following Thrombolysis?) Trial Study Group
- ↵The Strategies for Patency Enhancement in the Emergency Department (SPEED) Group. Randomized trial of abciximab with and without low-dose reteplase for acute myocardial infarction. Circulation 2000. In Press.
- Chesebro J.H,
- Knatterud G,
- Robert R,
- et al.
- Gibson C.M,
- Cannon C.P,
- Daley W.L,
- et al.
- Antman E.M,
- Giugliano C.P,
- Gibson C.M,
- et al.
- Ross A.M,
- Coyne K.S,
- Reiner J.S,
- et al.
- Fitzgerald D.J,
- Catella F,
- Roy L,
- FitzGerald G.A
- Brener S.J,
- Barr L.A,
- Burchenal J.E,
- et al.
- Montalescot G,
- Barragan P,
- Wittenberg O,
- et al.
- Antoniucci D,
- Santoro G.M,
- Bolognese L,
- Valenti R,
- Trapani M,
- Fazzini P.F
- Bode C,
- Smalling R.W,
- Berg G,
- et al.
- Matetzky S,
- Novikov M,
- Grugerg L,
- et al.