Early vs Late Administration of Glycoprotein IIb/IIIa Inhibitors in Primary Percutaneous Coronary Intervention of Acute ST-Segment Elevation Myocardial Infarction:  A Meta-analysis FREE

Primary percutaneous coronary intervention (PCI) is widely regarded as the reperfusion strategy of choice in ST-segment elevation myocardial infarction (STEMI),¹ but only a minority of STEMI patients present directly to PCI centers, while the vast majority present initially to their emergency ambulance service and/or to local hospitals and require transfer to a tertiary center for primary PCI. Despite the attendant delay, we confirmed in a recently published meta-analysis the superiority of transfer for primary PCI over local thrombolysis.² The absence or negligible use of glycoprotein IIb/IIIa (Gp IIb/IIIa) antagonists in the primary PCI arms of the studies of this meta-analysis may have reduced the magnitude of benefit observed with primary PCI. Indeed, 5 randomized studies have now shown a significant 30-day reduction in death or nonfatal MI or urgent revascularization with Gp IIb/IIIa antagonists,^3- 7 with a pooled analysis confirming further a significant 34% reduction of death or MI and a nonsignificant 26% reduction of mortality.⁸

The initial experience of exporting Gp IIb/IIIa inhibitors out of the catheterization laboratory was described in the Abciximab before Direct Angioplasty and Stenting in Myocardial Infarction Regarding Acute and Long-term Follow-up (ADMIRAL) randomized study in which, for a subset of patients, treatment occurred in the emergency department or even in the ambulance prior to arrival in the catheterization laboratory.³ The magnitude of the drug effect was increased dramatically in this subgroup compared with the rest of the population treated later in the catheterization laboratory.

More recently, several small angiographic studies have been performed to test the timing issue of Gp IIb/IIIa inhibitors in a randomized fashion.^9- 14 Half of these studies have been inconclusive on angiographic end points,^11,13- 14 and all have been too small to make meaningful conclusions regarding clinical end points. We therefore decided to perform a systematic meta-analysis.

Our primary aim was to compare, in STEMI, immediate treatment with Gp IIb/IIIa inhibitors prior to arrival in the catheterization laboratory vs a later administration of Gp IIb/IIIa inhibitors just before primary PCI. STEMI was defined according to the inclusion criteria of the trials concerned. It was recognized that there would be heterogeneity in the studies and that transfer times as well as patient management were likely to be variable. On the other hand, by combining the studies, even if case-specific data were not available, the statistical power to detect a difference in time effect might be substantially increased.

We conducted a MEDLINE and Cochrane Controlled Trials Register search of the literature to identify all randomized trials published in the last 10 years that compared early vs delayed administration of Gp IIb/IIIa inhibitors in STEMI. In addition, we searched for papers presented at major cardiac conferences. National and international colleagues as well as Gp IIb/IIIa inhibitor drug manufacturers were consulted. Finally, text and journal article bibliographies were hand searched. From these trials, specific "early vs late" studies were identified, defined as those in which randomization occurred outside the catheterization laboratory.

We restricted our meta-analysis to trials that performed a randomized comparison between early administration at the point of initial contact (emergency department or ambulance) and late administration (catheterization laboratory) of Gp IIb/IIIa inhibitors in STEMI. Outcome data had to be available on both culprit artery patency evaluated on admission by TIMI flow grades and mortality. Two authors (G.M., M.B.)independently reviewed abstracts or complete articles. Only 6 nonoverlapping studies met the inclusion criteria. Independent data extraction was performed by 2 reviewers and confirmed by consensus.

The primary angiographic end point was the combined TIMI grade 2 and 3 flows on the first angiogram, which defined open culprit artery on admission. TIMI grade 3 flow was also assessed separately. The primary clinical end point was all-cause mortality at longest follow-up. The composite ischemic end point at longest follow-up, as defined in each study, was also assessed separately.

The results from each trial were those obtained on an intention-to-treat basis. The meta-analysis was performed using the odds ratio (OR) as the parameter of efficacy with a fixed effect model (Mantel-Haenszel),¹⁵ with appropriate tests for association and heterogeneity. Finally, the number of patients needed to be treated to avoid 1 event (NNT) was calculated for each end point using the overall weighted risk difference (NNT = 1/[absolute risk difference]).

The P value for significance of association and heterogeneity tests was set at .05. The statistical analysis was performed using EasyMA software.¹⁵

Six trials were identified with a design including randomization to immediate vs delayed administration of Gp IIb/IIIa inhibitors for primary PCI, in which randomization took place outside the catheterization laboratory.^9- 14 The trial names, acronyms, patient characteristics, and details of the study groups are shown in Table 1.

Enrollment criteria of chest pain (<6 or <12 hours) with ST elevation or new left bundle-branch block on the electrocardiogram were used throughout (Table 1). Trial design was broadly similar in all studies; however, there were some differences regarding site of randomization: in 4 studies, randomization occurred in the emergency department of the PCI center^10- 12,14; in 1 study, randomization was performed at the scene of presentation outside the hospital and the drug was given during ambulance transportation¹³; and in the last study, patients could be enrolled and randomized in the ambulance, at referring hospitals, or in the emergency department of the PCI center but always before transfer to the catheterization laboratory.⁹ One of the 6 randomized studies was double-blind, placebo-controlled.⁹ The same dose regimen of tirofiban was used throughout the 3 tirofiban studies with a bolus of 10 µg/kg followed by a 0.15 µg/kg/min infusion,^9- 11 and the same dose regimen of abciximab was used throughout the 3 abciximab studies with an intravenous bolus of 0.25 mg/kg followed by a 0.125 µg/kg/min infusion (to a maximum of 10 µg/min).^12- 14

Mortality data were available at 1 month follow-up in 3 studies,^10- 11,14 at 6-month follow-up in 2 studies,^12- 13 and at 1-year follow-up in the last study.⁹ The use of a 30-day composite ischemic end point of death, reinfarction, or urgent revascularization was also consistent throughout the studies with a few differences between trials, as indicated in Table 1.

The proportion of patients with TIMI grade 2 or 3 flow was significantly higher with an early administration of Gp IIb/IIIa inhibitors in only 3 of the 6 trials.^9- 10,12 Combining all 6 trials together, early administration of Gp IIb/IIIa inhibitors was associated with a marked and highly significant increase in open infarct arteries as indicated by a TIMI grade 2 or 3 flow rate of 41.7% vs 29.8% with a late administration (OR, 1.69; 95% confidence interval [CI], 1.28-2.22; P<.001 [Figure 1]). There was no evidence of heterogeneity, and the overall NNT was 8 for this primary angiographic end point. Odd ratios were similar for the 2 tested drugs with a nonsignificant interaction test for the type of drug.

The proportion of patients with TIMI grade 3 flow on the initial angiogram was greatly increased in the early administration group (Figure 2). The overall NNT was 12 for achievement of TIMI grade 3 flow.

The mortality rates varied between trials, and none of the 6 trials showed a significant difference between the 2 study groups. Combining the trials, there was a 28% reduction in mortality in the early administration group (16/467 [3.4%]) compared with the late administration group (22/466 [4.7%]), which did not reach statistical significance (Figure 3). There was no evidence of heterogeneity, and the ORs were similar for the 2 tested drugs with a nonsignificant interaction test for the type of drug. A similar favorable trend was found in the group receiving GP IIb/IIIa inhibitors early for the composite ischemic end point (OR, 0.78; 95% CI, 0.51-1.20; P = .32) and for MI as individual component of this end point (OR = 0.73, 95% CI, 0.31-1.77; P = .64).

Because small trials are more prone to be affected by publication bias, we constructed funnel plots for both angiographic and clinical end points that exhibited a fairly symmetrical distribution and convergence toward the pooled effect when the weight of the trials increased, suggesting that publication or selection bias was unlikely.

The attendant delay for primary PCI may limit its clinical benefit over thrombolysis, especially when patients present within 3 hours after the onset of symptoms.^16- 17 It has been argued that "early pharmacologic facilitation" during the transfer period to the catheterization laboratory may improve outcome, although the efficacy of this strategy remains to date unproven. There is robust evidence for the superiority of Gp IIb/IIIa inhibition over placebo in primary PCI.^{3- 8,18- 21} However, the huge majority of patients recruited in Gp IIb/IIIa inhibition clinical trials received treatment relatively late, at the time of PCI.

In the ADMIRAL study, the analysis of the prespecified subgroup that received abciximab in the emergency department or in the ambulance showed better outcomes than the group of patients receiving the drug later, suggesting an advantage of "facilitation."³ However, to definitively demonstrate facilitating efficacy requires randomization against placebo, immediate administration of the study drug on presentation, and similar pharmacologic treatment in both study arms during and after PCI. The 6 studies we identified fulfilled these requirements.

The major finding of our meta-analysis is the demonstration of a probable "facilitating" effect of Gp IIb/IIIa inhibitors for improving angiographic patency of the culprit artery. TIMI flow on initial angiography is likely of clinical importance since it has been shown to be a major determinant of survival in primary PCI²² and a valid surrogate end point for mortality in thrombolytic trials.²³ The early administration of Gp IIb/IIIa inhibitors was also associated with consistent and favorable trends for mortality, reinfarction, and the composite ischemic end point, although these differences remained nonsignificant.

Short time to treatment is a well-known factor for clinical benefit with thrombolysis²⁴ and with primary PCI.^25- 28 It appears that the same time-to-treatment paradigm applies to Gp IIb/IIIa inhibitors when they are used in conjunction with primary PCI. This time factor in the drug effect may also explain some of the discrepancies observed between primary PCI trials that had different designs for the time of drug administration.^3,5,8

The current meta-analysis has a number of limitations, including inevitable clinical heterogeneity between trials and an overall sample size of only 931 patients, which is adequate for TIMI grade 2 or 3 flow (power of 97%) but insufficient for clinical outcomes. Sensitivity analyses were also performed and comparable results were obtained with a random effect model or when the largest study was excluded from the meta-analysis. Our analyses were also limited to short-term outcomes, but Gp IIb/IIIa inhibitor studies have often demonstrated increasing survival benefit with longer follow-up.^29- 30 In spite of the various limitations, however, the individual trials were in most cases underpowered and the combined analysis allows more robust conclusions.

Our data support strongly several hypotheses developed in recent years regarding what constitutes optimal myocardial reperfusion: first, the "open artery hypothesis"; second, the "dethrombotic" effect of Gp IIb/IIIa inhibitors^31- 32; third, the key role of time to treatment to improve outcomes; fourth, the concept of "facilitated" PCI, which still needs to be confirmed for clinical outcome; and fifth, the need to develop and/or consolidate prehospital/emergency department systems in the chain of care provided for STEMI patients. Whether more aggressive pharmacologic approaches using a full regimen of thrombolytics or reduced dose of lytics combined with Gp IIb/IIIa inhibitors can open more arteries more rapidly with finally improved clinical outcomes is so far unknown. Recent studies have suggested little impact on infarct size and clinical outcomes of these new facilitating strategies,³³ which now need to be tested in large-scale trials such as the Safety and Efficacy of a New Thrombolytic (ASSENT)-4 and Facilitated Intervention with Enhanced reperfusion Speed to Stop Events' (FINESSE) trials.

In conclusion, in STEMI patients treated with Gp IIb/IIIa inhibitors, there appears to be significant angiographic benefit with favorable trends for clinical outcome when treatment is started at first medical contact before transfer to the catheterization laboratory. Further evaluations in adequately powered large trials are awaited to confirm the clinical benefit of this strategy.