Author + information
- Received July 30, 2010
- Revision received September 28, 2010
- Accepted October 28, 2010
- Published online May 10, 2011.
- Jaroslaw Zalewski, MD⁎,
- Kris Bogaerts, PhD†,
- Walter Desmet, MD⁎,
- Peter Sinnaeve, MD⁎,
- Peter Berger, MD‡,
- Cindy Grines, MD§,
- Thierry Danays, MD∥,
- Paul Armstrong, MD¶ and
- Frans Van de Werf, MD⁎,⁎ ()
- ↵⁎Reprint requests and correspondence:
Prof. Frans Van de Werf, Department of Cardiovascular Medicine, Katholieke Universiteit Leuven, Herestraat 49, 3000 Leuven, Belgium
Objectives This study investigated the occurrence of intraluminal thrombus and its potential implications with facilitated percutaneous coronary interventions (fPCIs).
Background The effect of fPCI on the presence and consequences of intraluminal thrombus is unknown.
Methods Thrombolysis In Myocardial Infarction (TIMI) flow grade, frame count, and thrombus grade; distal embolization; and slow flow in the infarct-related artery were assessed in a blinded fashion on coronary angiograms in 1,342 patients from the ASSENT-4 PCI (Assessment of the Safety and Efficacy of a New Treatment Strategy With Percutaneous Coronary Intervention) trial. Residual TIMI thrombus grade ≥2 and/or distal embolization and/or slow flow, reflecting thrombus burden (TB), following PCI were correlated with ST-segment resolution, epicardial blood flow, and clinical outcome. The clinical composite endpoint was death, congestive heart failure, or shock.
Results In the fPCI group, more TIMI flow grade 2/3 in the infarct-related artery at the first angiogram (73.7% vs. 33.4%, p < 0.001) and a higher TB following PCI (19.7% vs. 13.4%, p = 0.002) were found in comparison with the primary PCI group. Post-PCI TIMI thrombus grade was significantly associated with ST-segment resolution (p < 0.001) and TIMI frame count (p < 0.0001) in both groups. In the fPCI group, the presence of post-PCI thrombus was associated with a significantly worse outcome at 90 days (clinical composite endpoint: 32.1% vs. 18.6%, p = 0.023). Multivariable logistic regression showed that facilitation with tenecteplase (p = 0.005) and TB (odds ratio: 2.43, 95% confidence interval: 1.30 to 4.51, p = 0.0052) were independent predictors of 90-day mortality.
Conclusions In ASSENT-4 PCI, despite more patency, residual TB was significantly higher in fPCI patients and was associated with less efficient tissue reperfusion and worse clinical outcomes. (A Trial Evaluating the Efficacy and Safety of Tenecteplase Together With Unfractionated Heparin Prior to Early Percutaneous Coronary Intervention [PCI] as Compared to Standard Primary PCI in Patients With Acute Myocardial Infarction [ASSENT-4 PCI]; NCT00168792)
- facilitated percutaneous coronary intervention
- primary percutaneous coronary intervention
- ST-segment elevation myocardial infarction
Primary percutaneous coronary intervention (pPCI) is the most effective reperfusion therapy (1) in patients with ST-segment elevation myocardial infarction (STEMI) but is inherently associated with an extra time delay (2). In an attempt to achieve reperfusion more quickly, facilitated percutaneous coronary intervention (fPCI) has been proposed and has been compared with pPCI in a number of trials. Neither facilitation with full-dose lytics, glycoprotein (GP) IIb/IIIa inhibitors, or both (3) improved clinical outcomes despite achieving higher patency rates documented by angiography. The reasons for this unexpected failure of fPCI remain unknown. It has been hypothesized that the performance of a percutaneous coronary intervention (PCI) in a prothrombotic environment created by lytic therapy might be responsible for thrombotic complications and, therefore, a worse clinical outcome (4). Whether lytic therapy and/or GP IIb/IIIa antagonists given before PCI significantly reduce the amount of thrombus is unknown. However, thrombi are associated with a high risk of distal embolization (DE), microvascular obstruction, and suboptimal myocardial reperfusion; therefore, they potentially mediate an unfavorable clinical outcome in STEMI patients (5-7). If fibrinolytic agents and/or GP IIb/IIIa antagonists indeed reduce the thrombus burden (TB), other mechanisms must be responsible for the unfavorable clinical outcomes.
We performed a blinded, detailed analysis of the coronary angiograms obtained in the ASSENT-4 PCI (Assessment of the Safety and Efficacy of a New Treatment Strategy With Percutaneous Coronary Intervention) study. We wanted to relate infarct-related thrombus burden with reperfusion efficiency and clinical outcomes in patients randomized to fPCI versus pPCI.
Details of the ASSENT-4 PCI trial have been published previously (8). Of the 1,667 patients enrolled, the angiograms of 1,431 patients were sent to the core angiography laboratory in Leuven. Of these, 89 were of insufficient quality to allow quantitative analyses. Therefore, the present substudy is limited to 1,342 patients with analyzable angiograms, all of whom had 90-day follow-up. Among these, 656 had been assigned to fPCI and 686 to pPCI.
All angiograms were assessed centrally without knowledge of the treatment assignment. Angiograms were analyzed for the determination of the infarct-related artery (IRA), the presence and size of the intraluminal thrombus, epicardial blood flow, DE, collateral blood flow, and the presence of a significant stenosis (>50%) based on visual inspection. Intracoronary thrombus was scored at baseline after fibrinolytic therapy, during PCI (intraprocedural), and after PCI (post-PCI) using the Thrombolysis In Myocardial Infarction (TIMI) thrombus grade (TTG) scoring system (9). The sequence with the largest thrombus during PCI was determined. Patency of the IRA and epicardial blood flow were evaluated by means of the TIMI flow grade scale (10) and TIMI frame count scale (11). We defined DE as a filling defect with an abrupt cutoff in 1 or more coronary branches distally to the culprit lesion. Slow epicardial blood flow was determined as a temporary or permanent decrease in the TIMI flow grade scale of at least 1 grade after balloon inflation or stent implantation without evidence of spasm or dissection. Presence of a residual thrombus (TTG ≥2) after the procedure or evidence of DE or slow flow during or immediately following the procedure was defined as a composite angiographic endpoint. This endpoint was considered to be the expression of TB and was calculated only once.
The ST-segment elevation resolution (STR) analysis performed in the ASSENT-4 PCI trial has previously been described (12). Electrocardiograms were available for analysis in 981 (73.1%) patients at 60 min and in 1,090 (81.2%) patients at 180 min.
The clinical endpoints were pre-specified in the protocol (8). The primary clinical composite endpoint (CCE) was a combination of death, congestive heart failure, or shock 90 days after randomization. Congestive heart failure and shock were centrally adjudicated by a clinical event committee unaware of the allocated treatment. The definition of reinfarction was adopted from previous ASSENT trials (13).
Statistical analyses were performed with SAS (version 9.2, SAS Institute, Inc., Cary, North Carolina) and R (version 2.9, R Foundation for Statistical Computing, Vienna, Austria). Baseline characteristics and angiographic results were described by mean and SD or median and quartiles (Q1 and Q3) for continuous variables, and numbers and percentages for categorical variables. Between treatment groups, comparisons were done by a t test, Wilcoxon rank sum test, chi-square test, or Fisher exact test, as appropriate.
The relation between the presence of post-PCI thrombus and mortality, and CCE and reinfarction at day 90 for each treatment group was verified by Fisher exact test.
Multivariable logistic regression analyses were used to determine independent correlates of residual thrombus, CCE, and 90-day mortality. To account for missing data in the covariates, a multiple imputation strategy (5 imputations) was applied. First-order terms were forced into the model. Second-order interactions with treatment were verified. A significance level of 0.05 was applied without correction for multiple testing.
Baseline clinical characteristics are shown in Table 1. Compared with the pPCI group, fPCI patients had a longer median time delay between randomization and first balloon inflation but not a significantly longer median time between symptom onset and first balloon inflation. Significantly more patients underwent an intervention in the pPCI group. A clopidogrel loading dose after the procedure and GP IIb/IIIa inhibitors both before and after PCI were more frequently administered to pPCI patients.
Significantly more patients in the fPCI group had an open IRA on the first contrast injection (Table 2). Thrombus aspiration and PCI were performed more frequently in patients assigned to pPCI. After PCI, the TIMI flow grade rates were similar.
Clinical outcomes are shown in Table 3. The CCE and reinfarction rates were higher in the fPCI group at both 30 and 90 days, and by 90 days there was also a tendency for excess death, congestive heart failure, and shock in the fPCI group.
Core laboratory angiographic measurements
After fibrinolytic therapy, significantly higher rates of TTG 0 to 3 were observed in the fPCI patients, and more frequent TTG 5 was present in the pPCI patients (Fig. 1). During the PCI procedure, more patent vessels without thrombus were found in the fPCI patients, whereas more arteries with medium-size thrombus (TTG 2/3) were present in pPCI group. After the procedure, no statistically significant differences in TTG were present. However, trends toward more DE (p = 0.055) and large (TTG ≥2) residual thrombus (p = 0.069) were observed in patients randomized to fPCI. The calculated TB was significantly greater in fPCI patients (19.7% vs. 13.4%, p = 0.002). The relationship between post-PCI TTG and TIMI frame counts was significant in fPCI and pPCI patients (p < 0.0001) but different between these 2 groups (p = 0.036) (Table 4).
Using a multivariable logistic regression model, we found independent pre-procedural determinants of large (TTG ≥2) residual thrombus following PCI: post-fibrinolytic therapy TTG ≥3, no PCI, lack of stent implantation, loading dose of clopidogrel/ticlopidine, tenecteplase administration, age, time from onset of symptoms to first balloon inflation, and lack of prior PCI (Table 5).
Patients assigned to fPCI had significantly higher STR at 60 min (median [Q1, Q3]: 38.6% [0, 68.8] vs. 28.6% [–6.3, 58.1], p = 0.002), whereas STR at 180 min was similar in the 2 groups (71.4% [47.9, 87.5] vs. 71.2% [50.0, 87.5], p = 0.95). There was a strong relationship (p < 0.0001) between post-fibrinolytic therapy and post-PCI TTGs and the amount of STR in both groups (Fig. 2).
Clinical outcomes versus thrombus burden
Clinical outcomes analyzed according to the presence (or suspicion) of thrombus (TTG 1 to 5) versus no thrombus (TTG 0) are shown in Figure 3. Compared with the pPCI population, in the fPCI group, post-PCI TTG 1 to 5 was associated with a significantly higher risk of CCE (p = 0.023) and TTG 0 with a higher risk of reinfarction (p = 0.020).
Multivariable logistic regression showed that age, loading dose of clopidogrel/ticlopidine, Killip class 3/4 on admission, facilitation with tenecteplase especially in women, thrombus burden, and anterior wall infarction were independent correlates of the 90-day mortality (Table 6).
Clinical trials have indicated that lytic-based fPCI is associated with more patent IRA immediately before PCI (3,8). However, this did not translate into a clinical benefit. Our study provides a possible explanation for this remarkable finding.
We evaluated the frequency and consequences of residual intracoronary thrombus following fibrinolysis versus no lytic therapy and the amount of thrombus before and after mechanical fragmentation with PCI. An angiographic composite endpoint consisting of residual thrombus after the procedure, slow flow, and distal embolization during and/or immediately following the completion of the procedure was prospectively defined.
As expected, fPCI was associated with less total IRA occlusions but with more thrombus in the open vessels. Although more clearance of thrombus from the epicardial vessel was observed during the PCI procedure in the fPCI arm, this was associated with a significantly higher TB.
The presence of thrombus was associated with less STR, slower epicardial flow, and with a worse clinical outcome in both groups, but more so in the fPCI patients. The latter findings might be explained by the fact that a thrombus resistant to fibrinolysis is more likely to cause distal obstruction after subsequent fragmentation during PCI, and therefore more likely to impair tissue reperfusion. We have found that a longer time interval from symptom onset to first balloon inflation is independently associated with a larger residual thrombus. These findings are in agreement with the observation that intracoronary thrombi removed with distal protection devices during primary PCI are more highly organized and rich in fibrin in patients with later presentation (14). Consistent with the results of previous studies (15), it was also found that prior PCI decreases the amount of residual thrombus. Age, loading with thienopyridine, cardiogenic shock or pulmonary edema at randomization, intracoronary TB, anterior infarction, and fPCI independently influenced 90-day mortality. Surprisingly, the presence of residual intracoronary thrombus was not associated with a statistically significantly higher risk of reinfarction in both groups.
The well-known prothrombotic effects of fibrinolytic agents (16), the suboptimal antithrombotic cotherapy in this trial and, as a consequence, the presence of residual thrombus resistant to lytic therapy might be responsible for the unexpected failure of facilitated PCI.
Our study has several limitations, the most important being its retrospective nature. However, all analyses were performed blinded to the allocated treatment. Angiograms were available for analysis in only 81% of the patients enrolled. However, baseline characteristics of this subpopulation were similar to those of the total population. Furthermore, activated clotting time values were not collected, and the use of thrombus aspiration was lower than would be expected today. Finally, grading the amount of thrombus is to a certain extent subjective.
Our results indicate that with the antithrombotic cotherapy given in ASSENT-4 PCI, pre-treatment with a full dose of a fibrin-specific fibrinolytic agent does not decrease thrombotic consequences in the immediate peri-PCI period and, in fact, was associated with a greater TB, less favorable reperfusion, and a worse clinical outcome. These findings stress the importance of optimal timing of PCI and of optimal antithrombotic cotherapy after fibrinolysis for STEMI.
Dr. Bogaerts is a statistical consultant for Boehringer Ingelheim. Dr. Berger has served as a consultant to AstraZeneca, Boehringer Ingelheim, Eli Lilly/Daiichi Sankyo, Medicure, Accumetrics, and Ortho McNeil (each for <$10,000); he has received research funding for Geisinger Clinic for studies on which he is the Principal Investigator from Thrombovision, Helena, Accumetrics, AstraZeneca, Haemoscope, The Medicines Company, and Corgenic/Aspirinworks (each for >$10,000); and he owns equity in Lumen, Inc. Dr. Danays is an employee of Boehringer Ingelheim. Drs. Armstrong and Van de Werf received research grants from Boehringer Ingelheim to perform the ASSENT-4 PCI study. All other authors have reported that they have no relationships to disclose.
- Abbreviations and Acronyms
- clinical composite endpoint
- distal embolization
- facilitated percutaneous coronary intervention
- infarct-related artery
- percutaneous coronary intervention
- primary percutaneous coronary intervention
- ST-segment elevation myocardial infarction
- ST-segment elevation resolution
- thrombus burden
- Thrombolysis In Myocardial Infarction
- TIMI thrombus grade
- Received July 30, 2010.
- Revision received September 28, 2010.
- Accepted October 28, 2010.
- American College of Cardiology Foundation
- Collet J.P.,
- Montalescot G.,
- Le May M.,
- Borentain M.,
- Gershlick A.
- Henriques J.P.,
- Zijlstra F.,
- Ottervanger J.P.,
- et al.
- Wu K.C.,
- Zerhouni E.A.,
- Judd R.M.,
- et al.
- Stone G.W.,
- Peterson M.A.,
- Lansky A.J.,
- Dangas G.,
- Mehran R.,
- Leon M.B.
- Assessment of the Safety and Efficacy of a New Treatment Strategy with Percutaneous Coronary Intervention (ASSENT-4 PCI) Investigators
- The TIMI IIIA Investigators
- The TIMI Study Group
- Gibson C.M.,
- Cannon C.P.,
- Daley W.L.,
- et al.
- McDonald M.A.,
- Fu Y.,
- Zeymer U.,
- et al.,
- ASSENT-4 PCI Investigators
- Beygui F.,
- Collet J.-P.,
- Nagaswami C.,
- Weisel J.W.,
- Montalescot G.