Author + information
- Received November 30, 2015
- Revision received January 25, 2016
- Accepted February 11, 2016
- Published online May 23, 2016.
- Asim M. Rafique, MDa,b,
- Piyush Nayyar, MDa,
- Tracy Y. Wang, MD, MHS, MScc,
- Roxana Mehran, MDd,
- Usman Baber, MDe,
- Peter B. Berger, MDf,
- Jonathan Tobis, MDb,
- Jesse Currier, MDb,g,
- Ravi H. Dave, MDb and
- Timothy D. Henry, MDh,∗ ()
- aDepartment of Medicine/Cardiology, Cedars-Sinai Heart Institute, Los Angeles, California
- bDepartment of Medicine/Cardiology, UCLA Medical Center, Los Angeles, California
- cDuke Clinical Research Institute, Durham, North Carolina
- dZena and Michael A. Wiener Cardiovascular Institute, Mount Sinai Hospital, New York, New York
- eDepartment of Medicine/Cardiology, Mount Sinai Hospital, New York, New York
- fNorthwell Health, New York, New York
- gDepartment of Medicine/Cardiology, VA Medical Center, Los Angeles, California
- hDivision of Cardiology, Cedars-Sinai Heart Institute, Los Angeles, California
- ↵∗Reprint requests and correspondence:
Dr. Timothy D. Henry, Cedars-Sinai Heart Institute, 127 South San Vicente Boulevardd A-3100, Los Angeles, California 90048.
Objectives The study sought to compare the clinical efficacy and safety of P2Y12 inhibitors in patients with ST-segment elevation myocardial infarction (STEMI) undergoing primary percutaneous intervention (PPCI).
Background Limited data exist regarding the comparative efficacy and safety of P2Y12 inhibitors in STEMI patients undergoing PPCI.
Methods Clinical trials enrolling STEMI patients were identified and relevant data was extracted. Major adverse cardiovascular events (MACE) were defined as the composite of all cause mortality, MI, and target vessel revascularization. Network meta-analysis was performed using Bayesian methods.
Results A total of 37 studies with 88,402 STEMI patients and 5,077 MACE were analyzed. Outcomes at 1 month (22 studies and 60,783 patients) suggest that prasugrel was associated with: lower MACE than clopidogrel (standard dose odds ratio [OR]: 0.59, 95% confidence interval [CI]: 0.50 to 0.69; high-dose OR: 0.60, 95% CI: 0.51 to 0.71; upstream OR: 0.79, 95% CI: 0.66 to 0.94), and ticagrelor (standard dose OR: 0.69, 95% CI: 0.56 to 0.84; upstream OR: 0.72, 95% CI: 0.50 to 1.05); lower mortality and MI than clopidogrel and standard ticagrelor; lower stroke risk than standard clopidogrel and standard or upstream ticagrelor; and lower stent thrombosis than standard or upstream clopidogrel. At 1-year (10 studies, n = 40,333) prasugrel was associated with lower mortality and MACE than other P2Y12 inhibitors. MACE was particularly lower with prasugrel in studies where patients received bivalirudin, drug-eluting stents, and but not glycoprotein IIb/IIIa inhibitor.
Conclusions In STEMI patients undergoing PPCI, prasugrel and ticagrelor are more efficacious than clopidogrel; in addition, prasugrel was superior to ticagrelor particularly in conjunction with bivalirudin and drug-eluting stents.
- P2Y12 inhibitors
- percutaneous coronary intervention
- ST-segment elevation myocardial infarction
Rapid, reliable, and potent inhibition of P2Y12 receptors in ST-segment elevation myocardial infarction (STEMI) patients undergoing primary percutaneous intervention (PPCI) is critical to reduce ischemic events including stent thrombosis (ST) (1–3). American College of Cardiology Foundation/American Heart Association 2013 STEMI guidelines give a class IB recommendation to clopidogrel, prasugrel, and ticagrelor (4). European Society of Cardiology 2012 STEMI guidelines recommend ticagrelor and prasugrel as preferred agents (Class I, Level of Evidence: B); and high-dose clopidogrel as an alternative agent (Class I, Level of Evidence: C) (5). Administration of a high loading dose of clopidogrel (600 mg) is associated with improved outcomes, however, delayed onset of action and variable platelet inhibition due to CYP2C19 polymorphisms remain a concern (6,7). Prasugrel and ticagrelor provide more rapid and potent platelet inhibition compared with clopidogrel, which translates into improved clinical efficacy in patients with acute coronary syndromes (ACS) (8,9). Post hoc analysis of STEMI patients in TRITON-TIMI-38 (Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel–Thrombolysis In Myocardial Infarction-38) and PLATO (Platelet Inhibition and Patient Outcomes) demonstrated improved clinical outcomes with prasugrel and ticagrelor compared to clopidogrel. It is clear that P2Y12 inhibition is a critical factor in the treatment of STEMI patients, yet there is no large randomized clinical trial comparing clinical outcomes in STEMI patients treated with various P2Y12 inhibitors. Therefore, a network meta-analysis was used to compare the outcomes in STEMI patients undergoing PPCI who were treated with P2Y12 inhibitors.
Network meta-analysis compares multiple treatments using both direct comparisons of interventions within trials and indirect comparisons across trials based on a common comparator.
Objectives, definitions, and study design
The primary objective of this analysis was to compare outcomes of P2Y12 inhibitors in STEMI patients undergoing PPCI. Clopidogrel (standard, high, and upstream) and ticagrelor (standard and upstream), and prasugrel administration was evaluated. A clopidogrel loading of 300 mg was considered standard, and 600 mg was considered high dose. Prasugrel 10 mg daily (60 mg loading dose) and ticagrelor 90 mg bid (180 mg loading dose) were considered standard dose. When data regarding loading dose was not specifically described, it was assumed that a standard dose had been administered.
Outcomes analyzed included: all cause mortality, cardiovascular mortality, MI, definite/probable ST, target vessel revascularization (TVR), stroke, and major/minor bleed at in-hospital (2 to 10 days), 1-month, and ≥1-year (12 to 36 months) follow-up. Major adverse cardiovascular events (MACE) were defined as all cause mortality, MI, or TVR. Studies were further stratified based on follow-up duration, bivalirudin use, glycoprotein IIb/IIIa inhibitor (GPI) use, transradial access, and bare-metal (BMS) versus drug-eluting (DES) stent use. Definite or probable ST was based on the academic research consortium definition (10). Major bleeding was based on Thrombolysis In Myocardial Infarction (TIMI) definitions (11).
Data source and study selection
A systematic literature review was performed using the following keywords: STEMI, ACS, MI, PPCI, P2Y12 inhibitors, thienopyridines, clopidogrel, prasugrel, and ticagrelor. Two investigators (A.M.R., P.N.) independently reviewed the studies. Conflicts were resolved by consensus and discussion with the senior author (T.H). Authors were contacted to obtain missing data where applicable.
Data extraction and quality assessment
Data were collected regarding age, sex, hypertension, dyslipidemia, diabetes mellitus, chronic kidney disease, smoking status, prior MI, prior PCI, prior coronary artery bypass grafts, anterior MI, left ventricular ejection fraction, % STEMI patients, % patients undergoing PPCI, follow-up interval, P2Y12 inhibitor dose and timing (periprocedural or upstream), bivalirudin use, GPI use, radial access, left main disease, stent type, or balloon angioplasty only and coronary artery bypass grafting, as well as MACE and bleeding outcomes.
Inclusion/exclusion criteria and study selection
Randomized and nonrandomized clinical trials enrolling STEMI patients undergoing PPCI with administration of periprocedural P2Y12 inhibitor were included. Trials where stratification of outcomes was not available by the type of P2Y12 inhibitor were excluded.
WinBUGS version 1.4.3 (MRC Biostatistics Unit, Cambridge, and Imperial College School of Medicine, London, United Kingdom) was used for mixed treatment comparison using the Bayesian methods (12). Dichotomous outcomes were compared using posterior median odds ratios and 95% Bayesian confidence intervals. Posterior distributions and network estimates were derived using Gibbs sampling via Markov Chain Monte Carlo simulation. Three chains were fit and checked convergence by assuring Monte Carlo error ≤5% of standard deviation of effect estimates and between-study variance. Convergence and lack of autocorrelation were confirmed after a 50,000 simulation burn-in phase using the Brooks-Gelman-Rubin test. A fixed-effect model was used to minimize the influence on effect size by smaller nonrandomized clinical trials and to reduce the impact of weak informative prior. Forest plots were used to illustrate the median effect estimate and corresponding 95% confidence interval.
Figure 1 shows the systematic search strategy. A total of 73 full text references were screened and 37 studies were included in the final review (Table 1). The study network plot is shown in Figure 2; further details of the network comparisons are available in Online Table 1.
Major characteristics of the studies including enrollment period (2003 to 2014), primary efficacy endpoints, bleeding definitions (TIMI definition in 19 studies) and follow-up intervals are provided in Table 1. The majority of the trials (n = 29) enrolled only STEMI patients; the other 8 trials reported clinical outcomes in the STEMI cohorts separately.
This analysis included 37 studies with 88,402 STEMI patients and 5,077 MACE events. Baseline demographics across the trials are shown in Online Table 2. The type, dose and timing of antiplatelet agents, and frequency of bivalirudin and GPI use are shown in Online Table 3. Periprocedural characteristics including prevalence of multivessel and left main disease, stent type, radial access and thrombectomy use, pre- and post-PCI TIMI flows are shown in Online Table 4.
In-hospital clinical efficacy end-points analysis included MACE (13 studies with 22,747), mortality (10 studies with 19,438 patients), cardiovascular mortality (3 studies with 1,374 patients), recurrent MI (9 studies with 17,736 patients), probable/definite ST (5 studies with 2,416 patients), and major bleeding (12 studies with 22,215). Prasugrel was associated with lower mortality and MACE than standard or high-dose clopidogrel; lower MACE than upstream clopidogrel and cangrelor but not compared to ticagrelor. Ticagrelor showed a trend to lower MACE than cangrelor. Cangrelor showed a trend to lower in-hospital mortality compared to standard clopidogrel (Online Tables 5 and 6).
Clinical efficacy end-points analyzed at 1-month included MACE (22 studies with n = 60,783), mortality (21 studies with n = 60,510), cardiovascular mortality (9 studies with n = 17,889), recurrent MI (18 studies with n = 51,629), probable/definite ST (13 studies with n = 29,907), and major bleeding (12 studies with n = 42,084) (Online Table 5). Prasugrel was associated with lower mortality, MI, and MACE than standard, high-dose, or upstream clopidogrel and standard ticagrelor; lower mortality and MACE than upstream ticagrelor; lower stroke risk than standard clopidogrel and standard or upstream ticagrelor; and lower ST than standard or upstream clopidogrel (Figures 3A to 3G). Standard dose ticagrelor showed a trend to lower 1-month mortality and MACE compared to standard dose clopidogrel. There was no significant difference in 1-month rates of cardiac mortality or major bleeding between prasugrel and other P2Y12 inhibitors (Figure 3H). Limited data was available for cangrelor (2 studies).
Clinical efficacy endpoints analyzed at 1-year included MACE (10 studies with 40,333 patients), mortality (10 studies with n = 41,766), cardiovascular mortality (3 studies with n = 13,742), recurrent MI (8 studies with n = 39,626), probable/definite ST (6 studies with n = 32,115), and major bleeding (7 studies with n = 23,489) (Online Table 5). Prasugrel was associated with lower mortality, MI, and MACE than standard or high-dose clopidogrel, and lower mortality and MACE than standard ticagrelor, a trend to lower MACE compared to upstream clopidogrel and standard ticagrelor. Prasugrel was associated with lower 1-year MI and TVR rate compared to standard or high dose clopidogrel; and lower ST risk compared to standard and upstream clopidogrel but not compared to ticagrelor (Figures 4A to 4G). Standard ticagrelor was also associated with significantly lower 1-year mortality, MI, ST, and MACE compared to standard or high-dose clopidogrel. No long-term follow-up data is available for cangrelor. There was no significant difference in 1-year rates of major bleeding between various P2Y12 inhibitors (Figure 4H).
Randomized trials and other subgroup analysis
Analyzing 14 randomized studies, prasugrel was associated with lower 1-month MACE than standard and high-dose clopidogrel, and standard ticagrelor. One-month MACE was further stratified based on: radial access in ≥50% of patients (5 studies), bivalirudin use in ≥50% of patients (10 studies), bivalirudin use in <50% of patients (12 studies), GPI use in ≥50% of patients (7 studies), GPI use in <50% of patients (15 studies), DES use in ≥50% of patients (11 studies), and bare-metal stent use in ≥50% of patients (11 studies). Prasugrel was associated with lower 1-month MACE than standard clopidogrel across all groups; and lower 1-month MACE than high-dose clopidogrel and standard ticagrelor in studies where ≥50% patients: received bivalirudin, did not receive GPIs, and received DES (Figures 5A to 5H).
This network meta-analysis is the largest study comparing the relative efficacy and safety profile of various P2Y12 inhibitors in STEMI patients undergoing PPCI. Despite the clinical importance and abundant observational data, there is a paucity of data from RCTs comparing P2Y12 inhibitors exclusively in STEMI patients. The principle findings of our network analysis are that at 1-month and 1-year follow-up, prasugrel was associated with a lower mortality, MI, and MACE rate than standard or high-dose clopidogrel; lower mortality and MACE than ticagrelor; and lower ST than standard or upstream clopidogrel. Ticagrelor was also associated with significantly lower 1-year mortality, MI, ST, and MACE than standard or high-dose clopidogrel. In patients receiving bivalirudin, DES, but not GP IIb/IIIa inhibitors; prasugrel was associated with lower 1-month MACE than clopidogrel and ticagrelor. There was no significant difference in major bleeding rates between various P2Y12 inhibitors in STEMI patients.
Higher platelet reactivity after clopidogrel loading is associated with larger intracoronary thrombus burden, worse post-PCI myocardial flow and perfusion (13). Genetic polymorphisms in CYP2C19, specifically *2 and *3 alleles result in reduced clopidogrel induced platelet inhibition (14). Slower onset can be partially overcome by higher dose or upstream administration of clopidogrel (15). In the TRITON-38 trial all patients received only a “standard” 300 mg loading dose of clopidogrel; in the PLATO trial, 36% of patients received a “high dose” 600 mg loading dose of clopidogrel. In the GRAVITAS (Gauging Responsiveness with a VerifyNow Assay—Impact on Thrombosis and Safety) trial high-dose clopidogrel did not reduce MACE in patients with high on-treatment reactivity (16). Upstream administration of a P2Y12 inhibitor may not be feasible for all STEMI patients. Prasugrel and ticagrelor achieve more rapid and greater platelet inhibition than clopidogrel (17–20). Following a 60-mg loading dose of prasugrel, 90% of ACS patients achieved 50% platelet inhibition at 1 h independent of CYP2C19 genotype (21). However, in STEMI patients, residual platelet reactivity after a loading dose of these new P2Y12 inhibitors is higher than in patients with stable coronary artery disease (22–24) likely due to poor drug absorption. In STEMI patients 2 h after loading dose ticagrelor achieved 12% platelet inhibition compared to 48% with prasugrel; and mean time to achieve PR <240 units was 3 ± 2 h with prasugrel compared to 5 ± 4 h in the ticagrelor group (25). A double loading dose of ticagrelor (360 mg) failed to achieve faster and more intense platelet inhibition (26). Therefore, a several-hour vulnerable window of suboptimal antithrombotic therapy exists in which STEMI patients are at risk of having ST. Cangrelor offers instant onset of action after intravenous administration; however, data is limited to 2 clinical trials and there are no studies comparing it to newer P2Y12 inhibitors.
Despite the limitations, these results point to better outcomes with prasugrel without significant increase in major bleeding. Bleeding is less frequent in STEMI patients due to younger age (on average 3 years) and fewer comorbidities. Lower bleeding rates with prasugrel may be related to selection bias due to Food and Drug Administration black box warning. In the PLATO trial mortality was higher than expected in the placebo arm, there was significant regional heterogeneity in clinical outcomes (46% MACE from 2 countries enrolling 21% patients) and outcomes were worse for U.S. patients, which was attributed to higher aspirin dose (27,28). A recent study reported that prasugrel but not ticagrelor was associated with lower 1-month mortality than clopidogrel in STEMI patients (29).
A randomized comparison of ticagrelor vs. prasugrel in patients with ACS and a planned invasive strategy (ISAR-REACT 5 [Intracoronary Stenting and Antithrombotic Regimen: Rapid Early Action for Coronary Treatment]) trial is underway. However, the relative merits of ticagrelor versus prasugrel in the treatment of ACS patients cannot necessarily be extended to STEMI patients. Our study highlights the need for a randomized clinical trial to compare various P2Y12 inhibitors in STEMI patients.
Our study includes the limitations of the original studies analyzed. Network meta-analysis assumes that patients enrolled in the various studies were sampled from the same theoretical population and that various drugs would have a consistent risk-benefit ratio across trials. Sources of heterogeneity include varying use of bivalirudin, GPI, DES, radial access, operator experience, door to balloon time, timing, loading dose and duration of P2Y12 inhibitor therapy, variable follow-up duration, and different definitions of clinical events across trials including MACE, bleeding, and recurrent MI. Outcomes with fewer events such as stroke, ST, and major bleeding may not be reliably estimated.
This network meta-analysis suggests that in STEMI patients undergoing PPCI, prasugrel is associated with better clinical outcomes than standard or high-dose clopidogrel at both 1-month and 1-year follow-up; ticagrelor is associated with better outcomes than standard or high-dose clopidogrel at 1-year; and prasugrel appears superior to standard ticagrelor at both 1 month and 1 year. Prasugrel is particularly more effective in patients receiving bivalirudin and DES.
WHAT IS KNOWN? P2Y12 receptor inhibition is critical to reduce ischemic events in patients undergoing PPCI. Despite this clinical relevance, there are no randomized trials comparing various P2Y12 inhibitors exclusively in STEMI patients.
WHAT IS NEW? In STEMI patients undergoing PPCI, prasugrel is associated with better clinical outcomes than standard or high-dose clopidogrel; ticagrelor is associated with better outcomes than standard or high-dose clopidogrel; and prasugrel appears superior to standard ticagrelor. The risk of major bleeding was not significantly different between various P2Y12 inhibitors.
WHAT IS NEXT? This study highlights the need for a randomized clinical trial to compare the clinical efficacy and safety of various P2Y12 inhibitors in STEMI patients.
The authors are thankful to the investigators of various clinical for sharing and clarifying the data, including Dr. Shahzad Adeel (United Kingdom), Dr. David J. Kurz (Switzerland), and Dr. Leonardo De Luca (Italy).
For supplemental tables, please see the online version of this article.
Dr. Wang has received research grants at her institution from AstraZeneca, Eli Lilly, Daiichi-Sankyo, and Bristol-Myers Squibb; and received consulting honoraria from AstraZeneca and Eli Lilly. Dr. Henry has served as a Steering Committee Member for TRANSLATE (supported by Eli Lily and Daiichi-Sankyo) and Artemis (supported by AstraZeneca). All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- acute coronary syndrome(s)
- drug-eluting stent(s)
- glycoprotein IIb/IIIa inhibitors
- major adverse cardiovascular event(s)
- primary percutaneous coronary intervention
- stent thrombosis
- ST-segment elevation myocardial infarction
- target vessel revascularization
- Received November 30, 2015.
- Revision received January 25, 2016.
- Accepted February 11, 2016.
- American College of Cardiology Foundation
- Spertus J.A.,
- Kettelkamp R.,
- Vance C.,
- et al.
- Airoldi F.,
- Colombo A.,
- Morici N.,
- et al.
- Bonaca M.P.,
- Steg P.G.,
- Feldman L.J.,
- et al.
- O'Gara P.T.,
- Kushner F.G.,
- Ascheim D.D.,
- et al.
- Steg P.G.,
- James S.K.,
- Atar D.,
- et al.
- Patti G.,
- Bárczi G.,
- Orlic D.,
- et al.
- Trenk D.,
- Hochholzer W.,
- Fromm M.F.,
- et al.
- Cutlip D.E.,
- Windecker S.,
- Mehran R.,
- et al.
- Bovill E.G.,
- Terrin M.L.,
- Stump D.C.,
- et al.
- Vavuranakis M.,
- Vrachatis D.A.,
- Papaioannou T.G.,
- et al.
- ↵FDA Drug Safety Communication: Reduced effectiveness of Plavix (clopidogrel) in patients who are poor metabloizers of the drug. March 12, 2010. Available at: http://www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafetyInformationforPatientsandProviders/ucm203888.htm. Accessed May 21, 2015.
- L'Allier P.L.,
- Ducrocq G.,
- Pranno N.,
- et al.,
- PREPAIR Study Investigators
- Gurbel P.A.,
- Bliden K.P.,
- Butler K.,
- et al.
- Alexopoulos D.,
- Galati A.,
- Xanthopoulou I.,
- et al.
- Angiolillo D.J.,
- Saucedo J.F.,
- Deraad R.,
- et al.,
- SWAP Investigators
- Parodi G.,
- Valenti R.,
- Bellandi B.,
- et al.
- Trenk D.,
- Stone G.W.,
- Gawaz M.,
- et al.
- Marciniak T. FDA Advisory Committee for Ticagrelor. Ticagrelor for acute coronary syndromes, NDA 22–433, changes from original version to amended version. July 28, 2010. Available at: http://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/CardiovascularandRenalDrugsAdvisoryCommittee/UCM220193.pdf. Accessed May 21, 2015.
- Alexopoulos D.,
- Xanthopoulou I.,
- Gkizas V.,
- et al.