Author + information
- Received January 28, 1998
- Revision received July 24, 1998
- Accepted August 20, 1998
- Published online December 1, 1998.
- Rabih R Azar, MD, MSra,* (, )
- Raymond G McKay, MD, FACCa,
- Paul D Thompson, MD, FACCa,
- Jeffrey A Hirst, MD, FACCa,
- Joseph F Mitchell, DO, FACCa,
- Daniel B Fram, MD, FACCa,
- David D Waters, MD, FACCa and
- Francis J Kiernan, MD, FACCa
- ↵*Address for correspondence: Rabih R. Azar, MD, Division of Cardiology, Beth-Israel Deaconess Medical Center, 330 Brookline Avenue, Boston, Massachusetts 02215
Objectives. The purpose of this study was to compare the outcome of primary percutaneous transluminal coronary angioplasty for acute myocardial infarction (MI) when performed with or without the platelet glycoprotein IIb/IIIa antibody, abciximab.
Background. Abciximab improves the outcome of angioplasty but the effect of abciximab in primary angioplasty has not been investigated.
Methods. Data were collected from a computerized database. Follow-up was by telephone or review of outpatient or hospital readmission records.
Results. A total of 182 consecutive patients were included; 103 received abciximab and 79 did not. The procedural success rate was 95% in the two groups. At 30-day follow-up, the composite event rate of unstable angina, reinfarction, target vessel revascularization and death from all causes was 13.5% in the group of patients who did not receive abciximab, 4% (p < 0.05) in the abciximab group and 2.4% (p < 0.05) in the subgroup of patients (n = 87) who completed the 12-h abciximab infusion. At the end of follow-up (mean 7 ± 4 months), the composite event rate was 32.4%, 17% (p < 0.05) and 13.1% (p < 0.01) in these three categories respectively. Abciximab bolus followed by a 12-h infusion was an independent predictor of event-free survival, in a Cox proportional hazards model (relative risk 0.49; 95% confidence interval 0.24 to 0.99; p < 0.05).
Conclusions. Abciximab given at the time of primary angioplasty may improve the short- and medium-term outcome of patients with acute MI, especially when a 12-h infusion is completed.
Platelet aggregation and activation play a key role in the thrombotic process resulting from plaque rupture in acute myocardial infarction (MI) (1)and during percutaneous transluminal coronary angioplasty (PTCA) (2,3). Trials with aspirin have confirmed a significant reduction in morbidity and mortality with antiplatelet therapy in these conditions (4–6). More potent platelet antagonists have recently been developed, such as the chimeric monoclonal antibody c7E3 or abciximab, directed against the platelet membrane glycoprotein (GP) IIb/IIIa receptor (7). Abciximab markedly inhibits platelet aggregation in a dose-dependent manner (8)and significantly reduces ischemic events in patients undergoing percutaneous coronary revascularization (9,10), particularly those with unstable angina or recent MI (9,11).
Primary angioplasty performed as soon as possible after the onset of symptoms is an efficient modality for the treatment of acute MI (12). However, despite an initial high success rate, it is still limited by recurrent ischemia that can lead to repeat catheterization before or soon after hospital discharge (12–14). Restenosis is also higher than after elective angioplasty (15,16). Thus patients with acute MI undergoing primary angioplasty represent a high risk group that may particularly benefit from c7E3 therapy given at the time of the procedure. This novel use of abciximab has not yet been fully investigated. The purpose of this study was to assess if the addition of c7E3 to primary angioplasty for acute MI improves the short- and medium-term outcomes of these patients.
The cardiac catheterization laboratory database at Hartford Hospital was reviewed from April 1996 to July 1997 to identify patients who underwent primary angioplasty for acute MI. For the purpose of the study, MI was defined as chest pain consistent with myocardial ischemia, with at least 1-mm ST segment elevation in two or more contiguous electrocardiographic (ECG) leads, or as ongoing chest pain refractory to medical therapy and associated with ST segment depression or left bundle branch block. Patients with postinfarction angina, failed thrombolytic therapy and angioplasty of a bypass graft were excluded. A total of 182 consecutive patients met these criteria; they were all included in the study.
All patients were initially treated with aspirin and intravenous heparin, given as a 5,000-U bolus followed by a continuous infusion, and were immediately taken to the cardiac catheterization laboratory. Patients also received nitrates unless they were hypotensive or in cardiogenic shock. The heparin dose was not adjusted to body weight, and an additional 5,000 to 10,000 U were given to all patients during angioplasty to achieve an activated clotting time above 300 s. Angioplasty was performed in the standard fashion. Intracoronary stenting with Palmaz-Schatz (Johnson and Johnson) stent was performed in 48% of the cases (51% of abciximab patients and 43% of control subjects; p = 0.32). Stent implantation was decided solely by the operator and was either planned (primary elective stenting), or placed for suboptimal results or as a bailout for complications. All patients who had a stent received ticlopidine, which was started as a single dose of 500 mg immediately after the procedure and continued at 250 mg twice a day (in addition to aspirin) for a period of 4 weeks. Use of abciximab was left to the discretion of the operator. Abciximab was given as a bolus of 0.25 mg/kg body weight, followed by a 12-h infusion at a rate of 10 μg/min. All sheaths were removed within the next 24 h, once the activated coagulation time dropped below 180 s, or earlier if local complications occurred.
Clinical data were collected while patients were hospitalized, and entered into the database. All data were verified by retrospective review of patient records. Postdischarge follow-up was obtained via telephone calls or by review of outpatient clinic charts. If a clinical event was reported, it was confirmed by review of hospitalization records. Follow-up was performed without knowledge of the procedural characteristics of the primary angioplasty (i.e., stent use, abciximab administration). Recurrent unstable angina after hospital discharge was defined as readmission with typical chest pain associated with either ischemic ECG findings, new myocardial perfusion abnormalities on nuclear imaging in cases where a baseline perfusion study was available or refractory chest pain requiring repeat coronary revascularization. Reinfarction before hospital discharge was defined as an increase in the activity of creatine kinase or its MB fraction of at least 50% from the previous trough, and remaining at least twice the upper limits of normal. Reinfarction occurring after discharge was defined as recurrence of typical chest pain associated with at least a twofold elevation in cardiac enzymes, or with new abnormal Q waves in two or more contiguous leads.
All coronary angiograms were reviewed by an interventional cardiologist without knowledge of abciximab administration or patient outcome. Mean diameter stenosis (visual estimation), Thrombolysis in Myocardial Infarction trial (TIMI) coronary blood flow grade (17)and thrombus score were determined before and after the procedure. The following thrombus score was used: 0 = no angiographically visible thrombus, 1 = possible thrombus, 2 = definite thrombus ≤3 mm in diameter and 3 = definite thrombus >3 mm in diameter.
Study end points
The primary end point of the study was a prespecified composite of any of the following during the total follow-up period: death from any cause, acute coronary syndrome (unstable angina or myocardial infarction) and target vessel revascularization. The later included patients who underwent emergent or elective repeat percutaneous revascularization or bypass surgery of the target vessel because of recurrent ischemia, but did not include patients with multivessel disease who had successful primary angioplasty and later underwent scheduled elective bypass surgery during which the target vessel was revascularized. A secondary end point was the same composite event rate at 30 days from primary angioplasty.
All continuous variables were expressed as mean value ± SD, and were compared with a two-tailed Student ttest. Categoric data were presented as absolute value and percent and were compared with a Fischer exact or a chi-square test. The primary end point of the trial was analyzed by evaluating the time to the first occurrence of any one of the components of the composite end point that occurred during the follow-up period. The results in the treatment groups are displayed as Kaplan–Meier event-free survival curves, which were compared by the log-rank method. Multivariate correlates of survival were analyzed with Cox model analysis. A p value <0.05 was considered statistically significant. Statistical tests were performed using commercially available software (SPSS 8.0, Chicago, IL).
Clinical and procedural characteristics
Of the 182 consecutive patients included in this analysis, 103 patients received abciximab and 79 did not (“no abciximab” group). In the abciximab group, the bolus of c7E3 was administered 5 to 10 min before crossing the lesion with the angioplasty guide wire in 79% of cases, and later during or immediately after the procedure in the remaining cases. Abciximab infusion was continued for 12 h in 87 patients (84%) but discontinued earlier in the remaining 16 patients because of need for early bypass surgery (three patients), and bleeding complications (13 patients). Mean activated clotting times were 343 ± 109 s and 376 ± 113 s (p < 0.05) in the abciximab and “no abciximab” groups respectively. Heparin was continued after the procedure in 29 (28%) abciximab patients (mean duration 26 ± 19 h) compared to 67 (85%) of patients who did not receive c7E3 (mean duration 24 ± 22 h).
Clinical characteristics (Table 1)were similar between the two groups, except that diabetes mellitus and hypertension were less prevalent in abciximab patients (9% and 38% vs. 24% and 57% respectively; p < 0.05 for both).
The total procedural acute success rate (TIMI III flow with ≤50% residual stenosis) was 95%. The procedural characteristics (Table 2)were similar between the two groups, except that abciximab patients had higher thrombus scores (1.46 ± 0.88 vs. 1.05 ± 0.6; p < 0.001) and lower activated clotting times measured before balloon dilatation (343 ± 109 vs. 376 ± 113 s; p < 0.05). After the procedure, 97% of patients had elevation in their cardiac enzymes more than twice the upper normal limit, confirming the diagnosis of acute myocardial infarction.
A mean follow-up of 7 ± 4 months was available on 174 patients (97% of the abciximab and 94% of the “no abciximab” patients; p = NS). Follow-up was obtained by telephone calls in 157 patients (90%) and by review of outpatient clinical charts in 17 patients (10%). All reported clinical events were confirmed by review of hospitalization records.
At 30 days, only four events (4%) occurred in the abciximab group compared to 10 (13.5%) in the “no abciximab” group (p < 0.05) (Table 3). This reduction was more important in patients who completed the 12 h infusion, as only two patients of the 84 on whom follow-up was available reached the composite end point (2.4%; p < 0.05).
Because more diabetics were present in the “no abciximab” group, and previous studies have associated diabetes mellitus with more complications following angioplasty (18,19), we measured the outcome in the two groups after exclusion of diabetic patients. The benefit of abciximab was maintained with a composite event rate of 3.3% at 30 days, in all patients who were treated with c7E3, compared to 15.8% in patients who did not receive abciximab (p < 0.05; Table 4).
Follow-up duration was shorter for abciximab patients (6 ± 3.6 vs. 8.3 ± 4 months; p < 0.001), reflecting the later introduction of abciximab into practice. Most of the events, however, occurred during the 2nd month after the index procedure; the average time to event was 2.4 ± 2.8 months in the abciximab group compared to 2.0 ± 1.8 months in patients who were not treated with abciximab (p = NS). During that period, a total of 17 (17%) abciximab patients reached the composite end point compared to 24 (32.4%) of patients in the “no abciximab” group (p < 0.05). This reduction was seen in all three components of the composite end point, but was statistically significant only in the recurrent unstable angina or reinfarction category (7% vs. 21.6%; p < 0.01) (Table 5). The composite event rate was lower in the subgroup of patients who completed the 12-h infusion (13.1%) (p < 0.01). Kaplan–Meier event-free survival curves during the total follow-up period are represented in Figure 1. Abciximab therapy resulted in better event-free survival, especially when a 12-h infusion was completed. Similar to the 30-day outcome, the protection conferred by GP IIb/IIIa antagonists persisted when diabetic subjects were excluded, with composite event rates of 13.2% in patients treated with abciximab compared to 31.6% (p < 0.05) in those who were not treated with abciximab (Table 4).
Because as many as 48% of our patients had at least one coronary stent implanted during angioplasty, we performed a subanalysis to assess the efficacy of the recommended abciximab regime (bolus + 12-h infusion) in primary angioplasty and stenting. Figure 2shows the Kaplan–Meier event-free survival curves for the four subgroups (angioplasty, angioplasty + stent, angioplasty + abciximab, angioplasty + stent + abciximab) for the total follow-up duration. Stent implantation alone did not provide additional benefit compared to the subgroup of patients who were not stented, but the combination of stenting and abciximab bolus plus 12-h infusion resulted in the best outcome.
Predictors of the medium-term outcome
In a univariate analysis, the predictors of the medium-term outcome were: gender, completion of 12-h abciximab infusion, number of vessels with coronary artery disease, use of abciximab, diabetes mellitus and cardiogenic shock. In a multivariate analysis using a Cox regression model that included all these variables, gender, single vessel coronary disease and completion of a 12-h abciximab infusion were the only independent predictors of event-free survival (Table 6). Abciximab use (with or without a 12-h infusion) was however, not a significant predictor of outcome in this multivariate analysis.
Major bleeding requiring blood transfusion occurred in 13 abciximab-treated patients (12.6%) and 5 (6.3%) of the nontreated patients (p = 0.24). Most bleeding complications occurred at the femoral access site, as groin or retroperitoneal hematomas in 12 of the 18 cases (67%). There was no intracranial hemorrhage, and none of the deaths was directly related to bleeding. Six patients also received blood transfusion after coronary bypass surgery (four patients who received abciximab and two patients who did not receive it) and after abdominal surgery (one patient who was not treated with abciximab and whose course was complicated with ischemic bowel). An additional 10 patients (five in each group) were transfused for a hematocrit less than 28% without a clinical source of blood loss. Thus, the total transfusion rate was 21.3% in patients treated with abciximab and 16.4% in those who were not (p = NS).
Despite an initial high success rate, primary angioplasty for acute myocardial infarction is associated with significant short- and long-term ischemic complications (12–16). In this series, the composite rate of death from all causes, recurrent unstable coronary syndromes and target vessel revascularization was 13.5% at 30 days, and 32.4% after a mean follow-up of 8.3 months, in the angioplasty group. Use of abciximab reduced the composite event rate to 4% at 30 days and to 17% after a mean follow-up of 6 months (70% and 47% reductions respectively). This benefit was more significant when the 12-h infusion was completed, with composite event rates of 2.4% and 13.1% during the short- and medium-term follow-up.
The present study is to our knowledge one of the first to directly address the use of abciximab in primary angioplasty for acute myocardial infarction. The results observed were similar to those recently reported in prospective trials using abciximab in high risk angioplasty. In the EPIC (Evaluation of c7E3 for the Prevention of Ischemic Complications) trial, abciximab bolus and infusion reduced the composite event rate by 35% at 30 days (9). The benefit persisted for 3 years in high risk patients (20). In EPIC, only 42 patients had primary angioplasty within 12 h of an acute myocardial infarction. Of those, 15 received abciximab bolus and infusion. Despite their limited number, a significant improvement in outcome was demonstrable at 6 months follow-up (21). Other than the EPIC post hoc subanalysis (which combined primary and rescue angioplasty together), trials on GP IIb/IIIa receptor inhibitors in primary angioplasty for acute myocardial infarction are still lacking.
Rationale for abciximab use in primary angioplasty
Multiple mechanisms may account for the efficacy of c7E3 in acute myocardial infarction. Most of the periprocedural complications of angioplasty are related to the thrombotic milieu of the unstable plaque. Several studies have demonstrated higher complication rates following intervention in thrombus-containing lesions (22)or in syndromes associated with intracoronary thrombosis (23). Platelet inhibition with aspirin in acute myocardial infarction and during angioplasty has long been established as an effective therapy, with reduction in acute thrombotic complications and in late clinical events (4–6). Abciximab is a more potent antiplatelet agent and may work in a similar but more effective fashion, converting the fissured or ruptured plaque and the endothelial surface from a platelet-reactive to a platelet-nonreactive surface. This potent antithrombotic action allows dissolution of preexisting clot and improvement of coronary blood flow, when used before (24)or after balloon inflation (rescue abciximab) (25)and may prevent distal embolization and the no-reflow phenomenon seen in primary angioplasty (26).
A 12-h infusion following abciximab bolus appears to be very important to obtain complete platelet inhibition and to achieve a significant antithrombotic effect. Similar to the EPIC trial (9), the best results in our series were obtained in patients who completed the 12-h infusion after an initial abciximab bolus. In the CAPTURE (c7E3 Fab Antiplatelet Therapy in Unstable Refractory Angina) trial, abciximab started 18–24 h before PTCA and continued for 1 h after the procedure, reduced myocardial infarction and death rates at 30 days, without benefit at 6 months (27). This may be related to interruption of therapy 1 h after angioplasty.
c7E3 and stents
Our results also suggest that combination of GP IIb/IIIa receptor blockers and stenting may be synergistic. Platelets are deposed heavily on the stent surface after its deployment, and their degree of activation is an independent predictor of acute stent thrombosis (28). In animal studies, acute stent thrombosis was less frequent in polymer-coated stents eluting platelet GP IIb/IIIa receptor antibody (29)and after systemic pretreatment with c7E3 (30). Human studies using these agents are underway, but recent trials showing the efficacy of ticlopidine have provided indirect evidence that more powerful antiplatelet therapy may be useful (31). Whereas the primary benefit of stenting is to prevent abrupt closure and restenosis, GP IIb/IIIa antagonists reduce subsequent unstable coronary syndromes, providing further improvement in the outcome of patients with acute myocardial infarction.
The transfusion rate in this series was higher than that previously reported for high risk angioplasty (9,12). The heparin dose was not adjusted to body weight, resulting in very high activated clotting times. Furthermore, heparin was continued after the procedure in 28% of patients who received abciximab and 85% of those who did not. These reasons may account for the high transfusion rate observed. Recent studies have emphasized that fewer bleeding complications occur with a less aggressive heparinization protocol aiming for activated clotting times in the 200-s range, and with early sheath removal after the procedure (10,27). The use of percutaneous puncture closure devices has also been shown to be a superior tool to manual pressure technique to achieve femoral hemostasis (32)and may be particularly useful after abciximab administration.
The low threshold for blood transfusion followed in our coronary care unit to maintain oxygen delivery for acute myocardial infarction patients might also have contributed to the high transfusion rate. A recent study found that 64% of blood transfusion after percutaneous coronary interventions were inappropriate and did not adhere to the American College of Physicians guidelines for transfusion (33).
Study limitations and clinical implications
This study is limited by its observational nature and by the fact that abciximab was given in a nonrandomized fashion. Diabetes mellitus and hypertension were more prevalent in the group of patients who did not receive abciximab. Although abciximab benefit persisted when diabetics were excluded from the analysis and the regime combining a bolus plus a 12-h infusion was independently associated with risk reduction in the regression model, a selection bias favoring abciximab patients cannot be completely excluded. On the other hand, despite abciximab patients having higher thrombus scores and in many instances receiving abciximab as “rescue” after one or more balloon inflations, abciximab therapy remained associated with a better outcome.
The mean follow-up duration was also slightly longer in the “no abciximab” group (8.4 vs. 6.1 months). This, however, would not account for the improved clinical results at 30 days, and is unlikely to affect the medium-term follow-up, because the majority of events occurred early during the 2nd month. This is confirmed by Kaplan–Meier event-free survival curves, where the reduction in events started early and persisted during the following months.
Stent implantation was also not randomized. Because of the observational nature of this study, we were not always able to define retrospectively the exact indication for stenting. It is likely, however, that operators would be unwilling to leave a suboptimal angioplasty result if stenting was technically feasible. It is probable that many patients had stents for less than optimal angioplasty results as well as bailout for complications. On the other hand, patients who did not receive a stent might have had “stent-like” balloon angioplasty results. The fact that stents did not improve outcome compared to balloon angioplasty may reflect a bias in placing stents after suboptimal or failed angioplasty attempts. Abciximab bolus and perfusion however, improved the outcome in the two groups of patients who did and did not receive a stent.
In conclusion, patients with acute myocardial infarction undergoing primary percutaneous revascularization with balloon angioplasty and/or stents represent a high risk group where platelet glycoprotein IIb/IIIa therapy may be particularly beneficial. This study suggests that abciximab may improve their short- and medium-term outcomes. The advantage of primary angioplasty over thrombolytic therapy (34)may thus be further improved or maintained over a longer duration when abciximab is used at the time of angioplasty.
The Randomized, Placebo-controlled Trial of Platelet Glycoprotein IIb/IIIa Blockade With Primary Angioplasty for Acute Myocardial Infarction (RAPPORT) was published after submission of this manuscript (Circulation 1998;98:734–41).
We thank Jeff Mather, MS, for his excellent statistical assistance, and Roger Mennett for his technical support.
- Evaluation of c7E3 for the Prevention of Ischemic Complications
- myocardial infarction
- Thrombolysis in Myocardial Infarction
- Received January 28, 1998.
- Revision received July 24, 1998.
- Accepted August 20, 1998.
- American College of Cardiology
- Frink R.J,
- Rooney P.A.J,
- Trowbridge J.O,
- Rose J.P
- Gasperetti C.M,
- Gonias S.L,
- Gimple L.W,
- Powers E.R
- Antiplatelet Trialists’ Collaboration
- Savage M.P,
- Goldberg S,
- Bove A.A,
- et al.
- Tcheng J.E,
- Ellis S.G,
- George B.S,
- et al.
- Lincoff A.M,
- Califf R.M,
- Anderson K.M,
- et al.
- Brodie B.R,
- Grines C.L,
- Ivanhoe R,
- et al.
- O’Neill W.W,
- Weintraub R,
- Grines C.L,
- et al.
- Sheehan F.H,
- Braunwald E,
- Canner P,
- Dodge H.T,
- Gore J,
- Van-Natta P
- Uusitupa M.I.J,
- Niskanen L.K,
- Siitonen O,
- Voutilainen E,
- Pyorala K
- Lefkovits J,
- Ivanhoe R.J,
- Califf R.M,
- et al.
- White C.J,
- Ramee S.R,
- Collins T.J,
- et al.
- ↵De Feyter PJ, Serruys PW, van den Brand M, Hugenholtz PG. Percutaneous transluminal coronary angioplasty for unstable angina. Am J Cardiol 1991;68:125–35B.
- Gold H.K,
- Garabedian H.D,
- Dinsmore R.E,
- et al.
- Muhlestein J.B,
- Karagounis L.A,
- Treehan S,
- Anderson J.L
- Neumann F.J,
- Gawaz M,
- Ott I,
- May A,
- Mössmer G,
- Schömig A
- Aggarwal R.K,
- Ireland D.C,
- Azrin M.A,
- Ezekowitz M.D,
- de Bono D.P,
- Gershlick A.H
- Makkar R.R,
- Litvack F,
- Eigler N.L,
- et al.
- Kussmaul W.G,
- Buchbinder M,
- Whitlow P.L,
- et al.