The CURRENT-OASIS 7 (Clopidogrel and Aspirin Optimal Dose Usage to Reduce Recurrent Events–Seventh Organization to Assess Strategies in Ischemic Symptoms) trial showed that a 7-day 150-mg maintenance dose (MD) clopidogrel could reduce cardiovascular events in subgroup patients who underwent percutaneous coronary intervention (PCI) compared with the 75 mg/day regimen, although whether prolonging the high MD clopidogrel (≥150 mg) treatment period to at least 4 weeks can reduce major adverse cardiac events in the patients who underwent PCI with and without high on-clopidogrel platelet reactivity (HPR) is still controversial. We searched Pubmed, Embase, and Cochrane Library from inception until September 2014 for randomized controlled trials that compared high versus standard MD clopidogrel in patients who underwent PCI. Seventeen trials involving 4,822 patients who underwent PCI included 2,879 patients who were allocated to the “HPR patients” subgroup and 1,943 to the “native patients” subgroup without paying attention to the clopidogrel reactivity before randomization. Compared with the standard therapy, the high MD clopidogrel was associated with a significant reduction in the risk of major adverse cardiac events (odds ratio [OR] 0.52, 95% confidence interval [CI] 0.39 to 0.71, p <0.0001) in patients who underwent PCI. The HPR patients subgroup was also benefited from such high MD treatment (OR 0.54, 95% CI 0.38 to 0.77, p = 0.0007). The observed benefits were mainly attributed to treatment-associated reduction in stent thrombosis (OR 0.43, 95% CI 0.23 to 0.78, p = 0.006) and target vessel revascularization (OR 0.38, 95% CI 0.20 to 0.74, p = 0.004). There was no difference in the rate of major/minor bleeding event between the high and standard MD group (OR 0.80, 95% CI 0.56 to 1.13, p = 0.21). In conclusion, the efficacy and safety of at least 4 weeks’ high MD clopidogrel is greater than that of standard therapy for patients who underwent PCI with and without HPR.
Recently, a 600-mg loading dose of clopidogrel followed by 75 mg/day maintenance dose (MD) on top of aspirin at least for 1 year has been recommended for patients scheduled for percutaneous coronary intervention (PCI). Compared with 300-mg loading dose, 600 mg clopidogrel achieves higher active metabolite plasma concentration and shows a faster onset of action, a superior and more prominent platelet inhibition, and improves short-term clinical outcomes in patients who underwent stent implantation mainly because of a reduction in periprocedural myocardial infarction (MI). Despite clopidogrel’s proved efficacy in reducing thrombotic events, subacute or late stent thrombosis (ST) still occurs in 1% to 2% of patients after PCI. Obviously, the benefits of high loading dose of clopidogrel are limited to the early phase of treatment, and patients may have inadequate antiplatelet effect resulting in undesired events during the maintenance phase, implying the need for a higher MD regimen. Besides, high on-clopidogrel platelet reactivity (HPR) has been associated with greater incidence of adverse events in patients who underwent PCI, and high MD clopidogrel has shown to enhance the potency of platelet inhibition and reduce HPR in low responder patients. Yet there are limited data on the clinical consequences of treatment with high MD clopidogrel for at least 4 weeks. Therefore, we performed a meta-analysis of randomized controlled trials (RCTs) comparing high versus standard MD clopidogrel in patients who underwent PCI with and without HPR.
Methods
Pubmed, Embase, and Cochrane Library were searched by 2 independent investigators (WM and YL) for relevant articles published before September 1, 2014. The search key words included the following terms: (“clopidogrel”), (“high,” “double,” or “150 mg”), and (“percutaneous coronary intervention,” “PCI,” “stent,” “angioplasty,” or “coronary intervention”), with the article types as “randomized controlled trial.” Language was restricted to English. References of original manuscripts and review articles were manually searched to avoid possible omission of eligible RCTs.
Inclusion criteria including (1) studies were RCTs; (2) participants were who underwent PCI; (3) clopidogrel treatment was high MD (≥150 mg) versus standard MD (75 mg) with the actual antiplatelet intervention time ≥4 weeks; (4) clinical events were reported in both arms; (5) particularly, studies including patients who underwent PCI with HPR diagnosed by one of the adenosine diphosphate–specific platelet function assays were defined as “HPR patients” subgroup and other studies including patients who underwent PCI without paying attention to the clopidogrel reactivity before randomization or the utilization of platelet function assay was only for evaluation of the pharmacodynamics were defined as “native patients” subgroup; and (6) according to the consensus recommendation, the accepted adenosine diphosphate–specific platelet function devices were light transmission aggregometry (LTA), VerifyNow-P2Y12, vasodilator-stimulated phosphoprotein phosphorylation, and the multiple electrode aggregometry (MEA). The exclusion criteria were (1) ongoing studies, (2) irretrievable data, and studies not reporting clinical outcomes at follow-up.
The primary efficacy end point was major adverse cardiac events (MACEs) defined as the combination of events including cardiovascular (CV) death, nonfatal MI, definite/probable ST, target vessel revascularization (TVR), and stroke. CV death and MI were used according to the study definition, whereas definite/probable ST was defined according to the Academic Research Consortium criteria. Secondary end point included each individual component of the primary end point (CV death, MI, ST, and TVR). The main safety end point was “major/minor bleeding,” defined as either according to the Thrombolysis In Myocardial Infarction major and minor criteria or as a “severe or moderate” bleeding according to the Global Utilization of Streptokinase and T-PA for Occluded Coronary Arteries (GUSTO) scale. “Any bleeding” defined as all the bleeding events including Thrombolysis In Myocardial Infarction bleeding, GUSTO bleeding, Bleeding Academic Research Consortium bleeding, and bleeding complications according to per protocol definition.
Data were extracted independently by 2 investigators (WM and YL). Discrepancies were resolved by consensus or a third investigator (JZ). For each RCT, the following data were abstracted: year of publication, location, study design, number of patients with clinical events, clinical setting, platelet function assay, definition of HPR, stent type, antiplatelet therapy strategy, duration of antiplatelet therapy and follow-up, efficacy, and safety outcomes. Study quality was appraised according to the Cochrane Collaboration method ; in detail, we evaluated information regarding random sequence generation, allocation concealment, blinding of participants and researchers, blinding of outcome assessment, incomplete outcome data, selective reporting, and other bias. Studies failed to meet at least 4 of the 7 standards were considered as being risk of high bias. No formal scoring system was used.
Analysis was performed using Review Manager 5.1 (The Cochrane Collaboration, Copenhagen, Denmark) and Stata 12.0 (StataCorp LP, College Station, Texas). Studies were grouped into 2 subgroups (HPR patients and native patients) based on the platelet reactivity status. The overall treatment effect was estimated by the pooled odds ratio (OR) with 95% confidence intervals (95% CIs) using a fixed-effects model (Mantel-Haenszel) or a random-effects model (DerSimonian-Laird), according to the value of heterogeneity. Heterogeneity was quantified with chi-square heterogeneity statistic (p value <0.05 for heterogeneous results), and inconsistency was assessed by means of I 2 , which quantifies the percentage of the variability that is because of heterogeneity rather than chance. I 2 values >25%, 50%, and 75% were assigned to low, moderate, and high degree of heterogeneity, respectively. Sensitivity analysis was performed by omitting each study in turn to assess the stability and consistency of the results. Subgroup analyses were conducted in terms of loading dose, follow-up, stent type, risk of bias, and proportion of acute coronary syndrome (ACS %). Begg’s funnel plot and Egger’s linear regression test were used to evaluate publication bias.
Results
The flowchart of study selection is presented in Figure 1 . From a total of 1,216 potentially relevant studies, 22 studies were finally retrieved for more detailed assessment and 17 RCTs were included in this meta-analysis, involving 4,822 patients (50.2% randomized to high MD clopidogrel and 49.8% randomized to standard MD clopidogrel). Among those 17 RCTs, 2,879 patients from 7 studies were classified as the HPR patients subgroup and 1,943 patients from the rest 10 studies were treated as native patients subgroup. Tables 1 to 3 lists the baseline characteristics and clopidogrel regimens of the 17 studies reported in full-text publication. Particularly, 4 studies were consistently identified as being at risk of high bias according to the Cochrane Collaboration methods mentioned before.
| Trial | Location | Design(RCT) | Patients | Clinical setting | Intervention duration | Follow up | Stent type | Risk of bias |
|---|---|---|---|---|---|---|---|---|
| GRAVITAS 2011 | USA and Canada | Multicenter | 1109/1105 | UA:29.5% NSTEMI: 10.1% STEMI:0.4% SCAD: 60% | 6 months | 6 months | DES | low |
| EFFICIENT 2012 | Turkey | Two center | 47/47 | SCAD: 100% | 1 month | 6 months | BMS | low |
| DOSER 2012 | Hungary | Single center | 36/38 | SCAD: 100% | 1 month | 12 months | BMS/DES | low |
| Samardzic 2014 | Croatia | Single center | 43/44 | UA: 16.1% NSTEMI: 22.9% STEMI: 60.9% | 1 month | 12 months | BMS/DES | high |
| Gremmel 2012 | Austria | Single center | 21/23 | SCAD:100% | 3 months | 3 months | BMS/DES | high |
| Tousek 2011 | Czech | Single center | 30/30 | UA/NSTEMI: 33.3% STEMI: 43.3% SCAD:23.3% | 1 month | 6 months | BMS/DES | high |
| Wang 2011 | China | Single center | 150/156 | SCAD: 80% NSTEMI: 20% | 12 months | 12 months | DES | low |
| Hazarbasanov2012 | Bulgaria | Single center | 97/95 | SCAD: 43% NSTEMI: 33% STEMI: 24% | 1 month | 6 months | BMS/DES | low |
| Beckerath 2007 | Germany | Single center | 31/29 | SCAD: 100% | 1 month | 1 month | BMS/DES | low |
| VASP-02 2008 | France | Multicenter | 58/62 | SCAD: 100% | 28 days | 28 days | BMS/DES | high |
| Han 2009 | China | Single center | 403/410 | UA/NSTEMI: 75% STEMI: 25% | 1 month | 1 month | DES | low |
| DOUBLE 2010 | Italy | Single center | 24/24 | STEMI: 100% | 1 month | 1 month | BMS | low |
| Roghani 2011 | Iran | Single center | 205/195 | SCAD: 100% | 1 month | 1 month | BMS/DES | low |
| Abuzahra 2008 | USA | Single center | 77/42 | SCAD: 56.3% UA: 20.2% NSTEMI: 18.5% STEMI: 5% | 1 month | 1 month | DES | low |
| Ren 2012 | China | Single center | 46/55 | SCAD: 100% | 1 month | 6 months | BMS/DES | low |
| Angiolillo 2008 | USA and Spain | Single center | 20/20 | SCAD: 100% | 1 month | 2 months | BMS/DES | low |
| ARMYDA-150 2011 | Italy | Single center | 25/25 | SCAD: 36% UA/NSTEMI: 64% | 1 month | 2 months | BMS/DES | low |
| Trial | Platelet function assay | Definition of HPR | Efficacy end points | Results high/standard | p value | Safety end points |
|---|---|---|---|---|---|---|
| GRAVITAS 2011 | VerifyNow-P2Y12 | PRU>230 | MACE(CV death/MI/ST) | 2.3% vs.2.3% | 0.97 | GUSTO bleeding |
| EFFICIENT 2012 | VerifyNow-P2Y12 | Inhibition<40% | MACE(CV death/ MI/ACS /TVR /ST) | 4.3% vs. 17% | 0.045 | TIMI major/minor bleeding |
| DOSER 2012 | LTA (5 μM ADP) | AGGmax>34% | MACE(CV death/MI/TVR) | 3.1% vs. 24.6% | 0.01 | TIMI major/minor bleeding |
| Samardzic 2014 | Multiplate (6.4 μM ADP) | AU>46 | MACE(CV death/MI/TVR /stroke) | 16.2% vs. 36.3% | 0.034 | BARC bleeding events |
| Gremmel 2012 | VerifyNow-P2Y12; VASP Multiplate(6.4 μM ADP) | PRU>235; PRI>50%; AU≥47 | ST, in-stent restenosis | 4.8% vs. 8.7% | NA | Bleeding complications, Major bleeding |
| Tousek 2011 | VerifyNow-P2Y12 | PRU>240 | MACE(CV death/MI/stroke) | 3% vs. 7% | 0.55 | TIMI major/minor bleeding |
| Wang 2011 | VASP | PRI>50% | MACE(CV death/MI/ACS/ST /TVR) | 9.3% vs. 19.2% | 0.008 | TIMI major/minor bleeding |
| Hazarbasanov2012 | Multiplate (6.4 μM ADP) | AU>46 | MACE(CV death/MI/ST/stroke) | 0% vs. 5.3% | 0.028 | TIMI major/minor bleeding |
| Beckerath 2007 | LTA(5 μM ADP); VerifyNow-P2Y12 | NA | MACT(MI/TVR) | 3.2% vs. 3.4% | NS | TIMI major/minor bleeding |
| VASP-02 2008 | VASP | PRI≥69% | MACE(CV death/MI/TVR /stroke) | 0% vs. 1.6% | 0.486 | Major/minor bleeding |
| Han 2009 | NA | NA | MACE(CV death/MI/TVR/ST) | 1.0% vs. 2.1% | NS | TIMI major/minor bleeding |
| DOUBLE 2010 | VASP; LTA; VerfyNow-P2Y12 Whole blood aggregometry; TEG | NA | Subacute stent thrombosis, ST | 0% vs. 4.2% | NA | Bleeding complications |
| Roghani 2011 | NA | NA | MACE(CV death/MI/ST) | 2.9% vs. 1.0% | NA | Bleeding complications |
| Abuzahra 2008 | LTA (20μmol/L ADP) | NA | MACE(CV death/MI/TVR) | 10.3% vs. 23.8% | 0.04 | TIMI major/minor bleeding |
| Ren 2012 | Thrombelastography | NA | MACE(CV death/MI/TVR) | 13.0% vs.20.2% | 0.486 | Bleeding complications |
| Angiolillo 2008 | LTA (5、20μmol/L ADP) | NA | MACE(death/MI/TVR/ST) | 0% vs. 0% | NS | Bleeding complications |
| ARMYDA-150 2011 | VerifyNow-P2Y12 | NA | MACE(death/MI/TVR/ST/stroke) | 0% vs. 0% | NS | Bleeding complications |
| Trial | Intervention (high dose vs. standard dose clopidogrel) |
|---|---|
| GRAVITAS 2011 | After PCI 600mgLD+150mg MD vs. placebo+75mg MD for 6months |
| EFFICIENT 2012 | 75mg per day for at least 3 days before PCI, after PCI 150mg MD vs. 75mg MD for 1 month, then 75mg for both for 5 months |
| DOSER 2012 | 600mg LD before PCI, 150mg MD vs. 75mg MD for 1 month, then 75mg for both for 11 months |
| Samardzic 2014 | 300mg/600mg LD before PCI, up to two additional 600mg LD and a range of 75-300mg MD to achieve and maintain optimal platelet reactivity (19–46 U) according to MEA vs. standard clopidogrel dose |
| Gremmel 2012 | 300/600mg LD before PCI, then 150mg MD vs. 75mg MD for 3 months |
| Tousek 2011 | 600mg LD before PCI, starting at 150mg MD with further dose increases according to VerifyNow to reach a PRU<240 vs. 75mg for 1 month |
| Wang 2011 | 300 mg LD before PCI, MD were increased in a stepwise manner according to VASP-PRI (up to 375 mg) vs. 75mg for 12months |
| Hazarbasanov 2012 | 300mg/600mg LD before PCI, after PCI 600mg reloading+150mg MD vs. 75mg MD for 1 month, then 75mg for both for 3months/12 months after BMS/DES respectively |
| Beckerath 2007 | 600 mg LD before PCI, then 150mg vs. 75mg MD for 1 month |
| VASP-02 2008 | 300mg/600mg before PCI, then 150mg vs. 75mg MD for 28 days |
| Han 2009 | 600 mg LD before PCI, then 150mg MD vs. 75mg MD for 1 month |
| DOUBLE 2010 | 300mg LD before PCI, then 150mg vs. 75mg MD for 1 month |
| Roghani 2011 | 600 mg LD before PCI, then 150mg MD vs. 75mg MD for 1 month |
| Abuzahra 2008 | 600mg/300mg LD before PCI, 75mg/bid MD vs. 75mg MD for 1 month, then 75mg for both for ≥1 year |
| Ren 2012 | 300mg LD before PCI, 150mg MD vs. 75 mg MD for 1 month, then 75mg for both for 5 months |
| Angiolillo 2008 | 600mg LD after PCI, 150mg MD vs. 75mg MD for 1 month, then 75mg for both for 1 month |
| ARMYDA-150 2011 | 600 mg LD before PCI, 75mg MD for 1 month, then 150mg vs. 75mg MD for another 1 month |
Quantitative synthesis showed that a high MD clopidogrel dose significantly reduced the risk for MACEs in comparison with a standard MD clopidogrel in patients who underwent PCI (3.1% vs 5.4%; OR 0.52, 95% CI 0.39 to 0.71, p <0.0001; I 2 = 1%). The number needed to treat to prevent 1 MACE was 42. The HPR patients subgroup was also benefited from such high MD clopidogrel treatment (3.6% vs 6.3%; OR 0.54, 95% CI 0.38 to 0.77, p = 0.0007; I 2 = 35%; Figure 2 ). Further analysis showed that the reduced risk of MACEs in the high MD clopidogrel group was mainly attributed to treatment-associated reduction in ST (0.6% vs 1.6%; OR 0.43, 95% CI 0.23 to 0.78, p = 0.006; I 2 = 0%) and TVR (1.5% vs 4.1%; OR 0.38, 95% CI 0.20 to 0.74, p = 0.004; I 2 = 0%). However, no significant differences were indicated in CV death (0.8% vs 1.4%; OR 0.61, 95% CI 0.34 to 1.08, p = 0.09; I 2 = 0%) and MI (1.5% vs 2.1%; OR 0.67, 95% CI 0.43 to 1.03, p = 0.07; I 2 = 0%; Figure 3 ).
For the outcome of major/minor bleeding reported in 10 studies, there was no statistical difference between the 2 MD groups (3.0% vs 3.7%; OR 0.80, 95% CI 0.56 to 1.13, p = 0.21; I 2 = 0%). This trend was maintained both in the HPR patients subgroup (p = 0.23) and native patients subgroup (p = 0.63; Figure 4 ). The risk of “any bleeding” was also comparable between the high MD group and the standard MD group (7.7% vs 7.0%; OR 1.11, 95% CI 0.89 to 1.38, p = 0.35; Figure 5 ).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
