Intracoronary Glycoprotein IIb/IIIa Inhibitors Improve Short-Term Mortality and Reinfarction in East Asian Patients with ST-Segment Elevation Myocardial Infarction after Thrombus Aspiration: A Meta-Analysis

Objective Intracoronary (IC) glycoprotein IIb/IIIa inhibitors (GPIs) after thrombus aspiration (TA) for patients with ST-segment elevation myocardial infarction (STEMI), as compared with percutaneous coronary interventions (PCI) alone, is still on debate. To address this issue, we performed a meta-analysis of results from prospective or randomized controlled trials on the topic. Methods We searched electronic and printed sources (up to June 20, 2016) according to the selection criteria. Data were abstraction and meta-analysis was performed using RevMan 5.3 software. Results The cohorts involved 14 articles describing 1,918 participants were included. The incidence of the short-term major adverse cardiac events (MACE) was significantly reduced with intracoronary GPIs after TA (odds ratio [OR]: 0.29; 95% confidence interval [CI]: 0.13 to 0.65, p=0.003). Benefits were noted for short-term mortality (OR: 0.31; 95% CI: 0.17 to 0.57, p=0.0002) and reinfarction (OR: 0.28; 95% CI: 0.10 to 0.78, p=0.01) in subjects who received intracoronary GPIs after TA. Moreover, the Thrombolysis in Myocardial Infarction (TIMI) trial grade 3 postprocedure (OR: 2.29; 95% CI: 1.72 to 3.04, P<0.00001) and complete ST-segment resolution (STR) rate (OR: 2.68; 95% CI: 1.85 to 3.87, P<0.00001) were both improved with intracoronary GPIs after TA. As a result, left ventricular ejection fraction (LVEF) at short-term follow-up showed a significant difference (OR: 7.33; 95% CI: 5.60 to 9.06, p<0.0001) in favor of the TA and intracoronary GPIs administration. Conclusions Our study demonstrates that intracoronary GPIs may have a synergistic effect with thrombus aspiration on short-term mortality, reinfarction, and cardiac functional recovery.


Introduction
Primary percutaneous coronary intervention (PCI) has become the preferred reperfusion modality for patients with acute ST-segment elevation myocardial infarction (STEMI) [1]. As we all know, the possibility of distal embolization of atherosclerotic plaque and thrombus with subsequent microvascular injury and increased infarct size during primary PCI is associated with adverse cardiovascular events [2]. Thrombus aspiration (TA) has the potential of reducing distal embolization and improving microvascular perfusion during primary PCI. Even though numerous international studies have been reported, there are still conflicting results on the clinical impact of thrombus aspiration during primary PCI [3,4]. Recent evidence from Routine Aspiration Thrombectomy With Percutaneous Coronary Intervention (PCI) Versus PCI Alone in Patients With ST-Segment Elevation Myocardial Infarction (STEMI) Undergoing Primary PCI (TOTAL) trial, the largest trial of thrombus aspiration in STEMI so far, suggested that routine thrombus aspiration, as compared with PCI alone, did not reduce the risk of major adverse cardiovascular events (MACE) within 180 days [5], consistent with those of Thrombus Aspiration during ST-Segment Elevation Myocardial Infarction (TASTE) trial [4] and the Intracoronary Abciximab and Aspiration Thrombectomy in Patients With Large Anterior Myocardial Infarction 2 Evidence-Based Complementary and Alternative Medicine (INFUSE-AMI) trial [6]. However, TA along with intracoronary (IC) glycoprotein IIb/IIIa inhibitors (GPIs) was associated with improved 30-day mortality in INFUSE-AMI trial [6], which suggested the synergistic effect of TA and GPIs might be attributed to improvement in clinical outcomes. On the other hand, some East Asian studies (especially in China) from the year 2008 to 2015 yielded conflicting or inconclusive results [6,7]. The reason for the discrepancy is unclear but may be related to low statistical power or difference among the ethnic groups studied.
In this meta-analysis, we aim to assess the effects of intracoronary GPIs after thrombus aspiration compared with PCI alone in STEMI patients from the year 2008 to 2015.

Data Sources and Searches.
We performed a systematic search for articles in the databases MEDLINE (via PubMed), EMBASE, and the Cochrane Library (Cochrane Central Register of Controlled Trials) up to June 20, 2016, using the following keywords: (thrombus aspiration) AND (intracoronary) AND {("abciximab"[Substance] or "abciximab"[All Fields]) or ("eptifibatide"[Substance] or "eptifibatide"[All Fields]) or ("tirofiban"[Substance] or "tirofiban"[All Fields]}. We also searched the China National Knowledge Internet database to retrieve relevant studies published in Chinese. We restricted the search to human studies but not language. Further articles were retrieved by a manual search of references from recent reviews and relevant published original studies. Studies were screened by reading the abstracts and titles and then selected after reading the full text.

Study Selection.
A study was selected if (1) the subjects were prospectively or randomly assigned to TA plus GPIs or PCI alone in a parallel-group design; (2) major adverse cardiac events were reported as outcomes; (3) GPIs were administrated by intracoronary during the procedure. We excluded studies that were cross-sectional or case-control designs. In case of duplicate publication, we chose the publication reporting on the primary analysis. The long-term clinical outcome was defined as more than three months, and the short-term clinical outcome was less than three months or in hospital.

Data Extraction and Quality
Assessment. Data were extracted independently by 2 investigators (Li R.J. and Hao P.P.) using a standardized extraction form and compared. Discrepancies were resolved by discussion with a third investigator (Chen Y.G.) and by referencing the original report. The grade of study quality was assessed as the previous metaanalysis [8,9].

Data
Analysis. RevMan 5.3, developed by the Cochrane Collaboration (http://tech.cochrane.org/revman, released on 13 June 2014), was used for the meta-analysis. Heterogeneity was tested with the chi-square and I 2 tests. Statistical significance was a 2-tailed P < 0.05. Results showing no significant differences were analyzed by the fixed effects model and those showing significant differences were analyzed by the DerSimonian-Laird random effect model. We also performed a sensitivity analysis to explore the robustness of our results. For MACE, mortality, and reinfarction, we evaluated publication bias using funnel plots and the fail-safe number with P < 0.05 (Nfs0.05), Nfs0.05 = (ΣZ/1.64) 2 − k, where k is the number of studies included in the meta-analysis. Any calculated Nfs value smaller than the number of observed studies indicated publication bias that might influence the meta-analysis results.
Seven studies depicting baseline characteristics were summarized in Table 2. For these studies, baseline characteristics were not significantly different between the two groups. Choi's study was an abstract from ANGIOPLASTY SUMMIT in 2009 [22], and we could not obtain accurate baseline characteristics although we have contacted the corresponding author. Other 6 studies described those baseline characteristics were balanced between two groups in papers, while no accurate data of baseline characteristics were obtained from [15][16][17][18][19][20].
In heterogeneity testing and sensitivity analysis, we also found no significant heterogeneity for studies reporting short-term MACE, death, and reinfarction, and exclusion of any single study did not alter the overall finding. The funnel plot assessing the publication bias is shown in Figure 6. We calculated the Nfs0.05 for MACE, death, and reinfarction. The comparative Nfs0.05 for MACE, death, and recurrent MI of the short term was 10.29, 14.93, and 2.97, whereas those of long term were -0.99, -0.99, and -0.82, which indicated publication bias that might influence the meta-analysis results.

Discussion
The main findings of the present meta-analysis are as follows: (1) a combination of thrombus aspiration and intracoronary GPIs seemed to be superior to PCI alone in terms  of enhancing myocardial perfusion, as assessed by postprocedural TIMI flow 3, and complete STR rate. Importantly, cardiac function at short-term follow-up, analyzed by LVEF, showed much better to be in the thrombus aspiration and intracoronary GPIs group over the PCI group.

Short-term reinfarction
(2) The incidence of short-term MACE was significantly reduced with intracoronary GPIs after thrombus aspiration, including death and reinfarction, whereas there was no trend towards better outcome in studies with long-term MACE. Thrombus aspiration during primary PCI is controversial, especially after the TOTAL trial which showed that routine thrombus aspiration did not reduce the risk of longterm MACE, as compared with PCI alone, and the findings are consistent with those of the INFUSE-AMI trial. However, a subgroup analysis of the INFUSE-AMI trial showed that thrombus aspiration plus intracoronary administration of GPIs improved myocardium perfusion and resulted in a better clinical prognosis [6]. A recent meta-analysis, summarizing the conflicting randomized controlled trials (RCTs) of comparing thrombus aspiration with the control arm, showed that thrombus aspiration along with GPIs is associated with improved 30-day mortality [23]. However, the authors performed metaregression and compared the studies with a higher proportion of glycoprotein IIb/IIIa inhibitor use and those with lower glycoprotein IIb/IIIa inhibitor use in the thrombus aspiration arm [23]. In contrast, we pooled the studies which directly compared thrombus aspiration plus GPIs with PCI alone, especially intracoronary administration. Our meta-analysis could demonstrate a reduced incidence of short-term MACE in the intracoronary GPIs after thrombus aspiration group, as compared to the PCI group. Further analysis showed that the benefit comes from reduced short-term mortality and reinfarction. Nonetheless, it has to be underlined that no benefit of long-term MACE was observed.

Evidence-Based Complementary and Alternative Medicine
Ahn et al. found that distal embolization was less likely to occur in patients undergoing intracoronary abciximab and thrombus aspiration as assessed by on-site measurement of the index of microcirculatory resistance (IMR) after primary PCI [24], which suggested that combination treatment using GPIs and thrombus aspiration may synergistically improve myocardial perfusion in patients with STEMI undergoing primary PCI. It was suggested that direct intracoronary injection of GPIs might be superior to its intravenous injection for improving myocardial perfusion due to creating a higher local concentration of around the coronary thrombus, and high local concentration may facilitate thrombus disaggregation and improve microvascular perfusion [24,25]. Our meta-analysis found that intracoronary GPIs after thrombus aspiration provided significant benefits in postprocedural TIMI flow grade 3, complete STR when compared with PCI alone. STR is a simple and reliable marker of effective myocardial reperfusion which correlates with cardiac functional recovery [26]. In the present meta-analysis, complete STR tends to be better in the intracoronary GPIs after thrombus aspiration group than that in the PCI alone group. This result coincided with the improvement in left ventricular function measured by LVEF and resulted in a trend towards better clinical outcomes.
A subgroup analysis of INFUSE-AMI [6] and a metaanalysis [27] suggest that a combination of thrombus aspiration and GPIs treatment is effective in decreasing infarct size and mortality as compared to each treatment alone or PCI alone. These findings are consistent with those of our meta-analysis. If most thrombotic materials are retrieved by thrombus aspiration catheter, GPIs could further dissolve residual thrombus and microemboli in the microvasculature.  Test for overall effect: Z = 0.47 (P = 0.64) Heterogeneity: ChＣ 2 = 1.97, df = 7 (P = 0.96); ) 2 = 0% Figure 5: Meta-analysis of the major bleeding or minor bleeding events before discharge between thrombus aspiration plus intracoronary GPIs and PCI alone. TA: thrombus aspiration; IC: intracoronary; GPIs: glycoprotein IIb/IIIa inhibitors; PCI: percutaneous coronary interventions. This might interpret why our results of the meta-analysis were different from other studies [4,5]. Notably, subgroup analysis showed that the type of aspiration catheter might influence the clinical outcomes and, in this aspect, ZEEK catheter and EXPORT catheter, which present a stronger aspiration capacity for moderate to large thrombi, were superior to Driver C.E. catheter [28]. Differences in aspiration capacity between ZEEK, EXPORT, and Diver 10 Evidence-Based Complementary and Alternative Medicine C.E. in this setting might influence the short-term outcome.

SE(log[OR])
We found there was no significant difference between the number of bleeding events in the intracoronary GPIs group and those in the control group, despite GPIs' antiplatelet activity and the risk of bleeding. This might be due to the type of GPIs-tirofiban, which was mostly used in our metaanalysis (92.9% used tirofiban). Tirofiban is a representative of small molecule glycoprotein IIb/IIIa inhibitor with reliable platelet inhibition and reversibility [29]. Tirofiban, the most applied glycoprotein IIb/IIIa inhibitor in East Asia now, was found in 13 included studies (Choi's study not mentioned due to the abstract in ANGIOPLASTY SUMMIT).
There were also some limitations in this meta-analysis. First, few studies were found in other countries of East Asia, and we could not perform analyses in other populations. Second, we calculated the Nfs0.05 to assess the publication bias and found the results of long-term MACE, death, and reinfarction were -0.99, -0.99, and -0.82, which indicated publication bias and might influence the meta-analysis results. Third, 92.9% studies in our meta-analysis used tirofiban, and subgroup analysis of different type of GPIs was unable to be performed.

Conclusions
Our study demonstrates that intracoronary use of glycoprotein IIb/IIIa inhibitors may have a synergistic effect with thrombus aspiration on short-term mortality, reinfarction, and cardiac functional recovery. Future RCTs are needed to assess the impact of concomitant glycoprotein IIb/IIIa inhibitors with thrombus aspiration on the long-term outcomes of patients with STEMI.

Data Availability
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.