Article Text
Abstract
Objective The clinical evidence regarding the influence of tailored antiplatelet strategy on adverse outcomes has been controversial. The aim of the study was to evaluate the significance of tailored antiplatelet therapy with respect to clinical adverse events in antiplatelet-resistant patients.
Methods Randomised studies that assess clinical relevance of personalised antiplatelet treatment in antiplatelet-resistant patients were identified through a literature search: PubMed, EMBASE, Web of Science and the Cochrane Library. The primary endpoint was the composite of death from any cause and stent thrombosis. All total clinical adverse events and bleeding complications were evaluated.
Results Data were combined across seven randomised studies comprising 12 048 subjects, of whom 3738 (31.0%) were found to be antiplatelet-resistant. Antiplatelet-resistant patients provided with tailored antiplatelet therapy showed less risk of death or stent thrombosis than those assigned conventional antiplatelet treatment (0.5% vs 2.2%; OR (95% CI) 0.25 (0.13 to 0.49), p<0.0001). A significant benefit in terms of total adverse event risk reduction was observed during follow-up for tailored vs conventional antiplatelet therapy (5.5% vs 10.0%; OR (95% CI) 0.40 (0.20 to 0.77), p=0.006). No statistical difference in bleeding complications was observed between these two groups (p=0.08).
Conclusions In the study, personalised antiplatelet treatment for antiplatelet resistance was found to be associated with less occurrence of death or stent thrombosis and the less risk of total clinical adverse events than conventional treatment, without increasing the risk of bleeding complications.
Statistics from Altmetric.com
Introduction
Antiplatelet drugs, such as aspirin and clopidogrel, are widely used in the primary and secondary prevention of cardiovascular disease.1 However, not all patients profit to the same extent. This can be partly explained by the reduced efficacy of antiplatelet agents. The phenomenon is called antiplatelet resistance, non-responsiveness, low responsiveness or high on-treatment platelet reactivity (HTPR) when addressed biochemically as failure to inhibit platelet function ex vivo.2–4 It has been demonstrated that patients who are resistant to aspirin or clopidogrel are at a greater risk of atherothrombotic events during the follow-up than aspirin-sensitive or clopidogrel-sensitive patients, respectively.5–11
However, one-size-fits-all still seems to be the current principle regarding antiplatelet treatment in clinical cardiology, and the individual response to antiplatelet agents has been ignored for a long time. Currently, only a few randomised studies have assessed the efficacy of tailored antiplatelet therapy on clinical adverse outcomes against treatment involving conventional aspirin and clopidogrel treatment in antiplatelet-resistant patients.12–18 However, it remains controversial whether antiplatelet-resistant patients benefit from a personalised strategy. Here, we performed a systematic review and meta-analysis of randomised trials to evaluate the clinical impact of tailored antiplatelet treatment relative to conventional antiplatelet treatment in antiplatelet-resistant patients.
Methods
Search strategy
Four electronic databases were searched using a combined text word and MeSH search strategy of the terms: PubMed (from 1809 to May 2013), EMBASE (from 1966 to May 2013), Web of Science (from 2003 to May 2013), and the Cochrane Library (from 1800 to May 2013). Relevant keywords related to antiplatelet treatment (“antiplatelet,” “anti-platelet,” “aspirin,” “acetyl salicylic acid,” “acetylsalicylic acid,” “clopidogrel,” and “plavix”) were used in combination with words related to tailored therapy (“tailor*,” “personal*,” “optimi*,” “optimal,” and “individual*”), to adverse events (“event*,” “consequence*,” “outcome*,” “recurren*,” and “prognos*”) and to study design (“random*”). The reference lists of relevant articles and reviews, editorials and letters on this topic were also reviewed. No language restrictions were enforced.
Selection criteria
Two investigators (JL and ZJ) independently assessed studies at the title and/or at the abstract level, and in duplicate again reviewed the full text articles for eligibility, with divergences resolved by a third investigator (JJ or LH). Eligible trials had to meet the following criteria for inclusion: (1) patients on antiplatelet treatment; (2) clear definition of antiplatelet resistance based on laboratory platelet function test; (3) classification of screened patients as antiplatelet-resistant or antiplatelet-sensitive group; (4) resistant patients were allocated 1:1 randomly to tailored or conventional antiplatelet therapy; (5) detailed treatment strategies for tailored or conventional group were declared, respectively; (6) a measure of clinical adverse events was used in both treating groups; and (7) data regarding occurrence rates of adverse events were reported. Exclusion criteria were as follows: (1) non-randomised trials; (2) irretrievable or duplicated data; and (3) ongoing trials.
Definitions of clinical outcomes
The primary end point was the composite of death from any cause and stent thrombosis. All total clinical adverse events were evaluated. These included periprocedural myonecrosis, periprocedural myocardial infarction (MI), acute coronary syndrome (ACS), revascularisation, stroke and rehospitalisation for ischaemic events in addition to death and stent thrombosis. The end points were defined according to the individual study protocols. Bleeding events of safety concern were reported according to the Thrombolysis In MI (TIMI) criteria or the Global Utilization of Streptokinase and Tissue plasminogen activator for Occluded coronary arteries (GUSTO) bleeding classification.19
Quality assessment
Freedom from bias was evaluated in accordance with the Cochrane Collaboration method based on the following methodological items: selection bias (random sequence generation and allocation concealment), performance bias (blinding of participants and personnel), detection bias (blinding of outcome assessment), attribution bias (incomplete outcome data), reporting bias (selective reporting) and other bias (including topic-specific and design-specific).
Data extraction
Data were extracted by two reviewers (JL and ZJ) independently and in duplicate, and disagreements were resolved by arbitration after discussion. The publication information of each study, baseline demographics, use of antiplatelet agents, detection of antiplatelet resistance, randomisation of resistant patients, treatment strategies, duration of follow-up and record of clinical adverse events and bleeding complications were extracted. Authors of the papers were individually contacted by email when the data were unclear or to obtain additional data.
Statistical analysis
Continuous variables are presented as mean±SD. Categorical variables are expressed as percentage or frequency. The κ-statistic was used to assess agreement between reviewers for study selection. The Cochrane Q statistic and the I2 statistic were used to assess heterogeneity across studies. The effects of tailored and conventional antiplatelet treatment on the composite of death or stent thrombosis were pooled using a fixed-effects model performed according to the Mantel and Haenszel method. The effects on clinical adverse events were combined using a random-effects model performed according to the DerSimonian and Laird method. The pooled risk of bleeding in both groups was calculated using a fixed-effects model. The measure of effect was reported by ORs with 95% CIs.
Subgroup analysis was conducted to identify possible sources of significant heterogeneity. The differences between the subgroups were determined using interaction tests.
Sensitivity analyses were used to assess the robustness of these results by removing each included study at the same time to assess and evaluate the remaining overall ORs of clinical adverse events.
Visual inspection of funnel plot and Egger's test were used to assess the presence of publication bias by plotting the SE against the log OR. A value of p<0.05 (two-sided) was considered to be statistically significant.
All statistical analysis was performed using Review Manager (RevMan) software (V.5.2) of the Cochrane Collaboration and Stata software (V.8.0, Stata Corporation, College Station, Texas, USA).
Results
Search results
Of 1721 citations identified in database search, shortlisted citations were retrieved and checked at the title/abstract level, excluding 1698 papers. We evaluated full texts for the remaining 23 studies for compliance to inclusion and exclusion criteria and finally identified 7 eligible randomised studies, of which 5 were prospective, 4 were multicentre, and 2 were double blinded (figure 1). The κ value for agreement for study selection between the two reviewers was 0.92.
Characteristics of eligible studies
General features of included studies are shown in online supplementary table S1. Across the seven selected randomised studies, a total of 3738 out of 12 048 (31.0%) patients undergoing percutaneous coronary intervention (PCI) were found to be antiplatelet resistant. Of these, five studies were conducted in Europe. All patients took aspirin and clopidogrel before and after determination of antiplatelet resistance. Flow cytometry for vasodilator-stimulated phosphoprotein index (n=2), light transmittance aggregometry (n=1) and VerifyNow P2Y12 assay (n=4) were used as platelet function tests. Clopidogrel resistance was assessed in all seven studies, aspirin resistance in only one study. Definitions of clopidogrel resistance differed according to various methodologies and study groups. All antiplatelet-resistant patients were 1:1 randomly assigned to a strategy of adjustment of antiplatelet drugs (tailored treatment group) or to a strategy of conventional treatment (conventional treatment group), and their characteristics and events contributed by each study to this analysis are shown in online supplementary table S2. Use of glycoprotein IIb/IIIa (GPIIb/IIIa) antagonist (abciximab and tirofiban) (n=3), another P2Y12 antagonist (prasugrel) (n=1), and intensified therapy of clopidogrel (n=3) were reported as adjusted strategies for tailored antiplatelet treatment, and this means that three of the seven studies included here involved the use of GP IIb/IIIa for antiplatelet resistant patients, and four involved tailoring of oral antiplatelet regimes. The treated patients in tailored and conventional groups were followed up for 1 month (n=5) or 6 months (n=2). Study quality and risk of bias were variable, and they are detailed in online supplementary table S3.
Clinical adverse events and bleeding complications
Death or stent thrombosis occurred in 50 patients (1.3%), and the rate of occurrence of total clinical adverse events (death, stent thrombosis, periprocedural MI, ACS, revascularisation, stroke and rehospitalisation for ischaemic events,) was 7.7% (n=289). Antiplatelet resistant patients assigned to tailored antiplatelet therapy showed less risk of incidence of death or stent thrombosis than those assigned to conventional antiplatelet treatment (0.5% vs 2.2%; OR (95% CI) 0.25 (0.13 to 0.49), p<0.0001; I2=0%, p for heterogeneity=0.45) (figure 2). Furthermore, a significant benefit in terms of adverse event risk reduction in the follow-up was found with tailored vs conventional antiplatelet therapy (5.5% vs 10.0%; OR (95% CI) 0.40 (0.20 to 0.77), p=0.006) (figure 3), but there was marked statistical heterogeneity among these studies (I2=73%, p=0.001). The overall pooled effect estimate analysis showed no statistical difference in bleeding complications in antiplatelet-resistant patients randomly assigned to receive tailored and conventional antiplatelet arms (8.4% vs 6.9%; OR (95% CI) 1.24 (0.97 to 1.59), p=0.08; I2=0%, p for heterogeneity=0.76; figure 4).
Subgroup analysis and potential sources of heterogeneity
A list of factors that may have contributed to clinical and methodological variations is shown in online supplementary figure S1. Subgroup analysis was carried out to further evaluate possible sources of heterogeneity based on these factors. A statistically significant difference in effect of tailored antiplatelet therapy was observed among subgroups according to the country of origin of the centre or centres (p=0.004), method for detecting antiplatelet resistance (p=0.004) or duration of follow-up (p=0.018), as detailed in figure 5 but not according to prospective design (p=0.55), strategy of tailored antiplatelet treatment (p=0.67) or method of administration of the agents (p=0.39), as detailed in online supplementary figure S2.
Sensitivity analysis
Sensitivity was analysed by excluding one included study at a time. It showed each individual study to be consistent with the direction and size of the overall ORs of clinical adverse events (all p<0.05) (table 1 and online supplementary figure S3).
Publication bias
A funnel plot of effect size versus SE was constructed according to total clinical adverse event rates in tailored and conventional antiplatelet groups, which visually seemed to be slightly asymmetrical (see online supplementary figure S4), and Begg's test reported negative publication bias (p=0.548).
Discussion
This is the first systematic review and meta-analysis of the literatures evaluating the clinical significance of tailored antiplatelet therapy in antiplatelet-resistant patients. The principal finding of the present study is that tailored antiplatelet treatment for antiplatelet resistance, compared with conventional antiplatelet therapy, is associated with significant reduction in the occurrence of death or stent thrombosis and total clinical adverse events (death from any cause, stent thrombosis, periprocedural myonecrosis, periprocedural MI, ACS, revascularisation, stroke or rehospitalisation), and not to increase bleeding complications.
Mechanisms and measurements of antiplatelet resistance
Laboratory aspirin resistance or HTPR is defined as biochemical failure of the antiplatelet agent to inhibit tests of platelet function ex vivo. The mechanisms responsible for antiplatelet resistance include insufficient dose, poor compliance, related gene polymorphisms, baseline platelet hyperactivity and accelerated platelet turnover.2–4 A myriad of tests are currently available to assess inhibition of platelet function induced by antiplatelet agents.2–4 Aspirin resistance can be diagnosed by measurement of platelet thromboxane A2 (TXA2) production or of thromboxane-dependent platelet function.2 ,3 TXA2 production can be determined by measuring the concentrations of stable metabolites of TXA2, such as thromboxane B2 in the blood and 11-dehydro-thromboxane B2 in the urine. Tests that depend on platelet thromboxane production include agonist-induced platelet aggregation, which can be measured by light transmission aggregometry, electrical impedance aggregometry or platelet function analysis-100, VerifyNow aspirin and bleeding time. Clopidogrel resistance can be assessed using adenosine diphosphate (ADP)-induced maximum platelet aggregation through light transmittance aggregometry, flow cytometric measurement of the expression of vasodilator-stimulated phosphoprotein and VerifyNow P2Y12 receptor assay.3 ,4 The consensus statement enables clinicians to use thresholds of platelet reactivity that have a clinical value and to differentiate patients with high and low risks of adverse events. However, the controversy over what cut-off value to use has lasted years. There are two main reasons why no definite cut-off value has been identified: (1) There is no standard laboratory method that can simulate the in vivo platelet response to antiplatelet therapy, and the cut-off values suitable for use with different methods can vary. (2) Not all platelet function tests are equally effective in measuring aspirin's antiplatelet effect.
Antiplatelet resistance and its clinical significance
Although little consistency exists about how to determine which measure and definitions to use to identify patients who seem resistant to aspirin or clopidogrel in individual studies, there has already been definite evidence from systematic reviews and meta-analyses that patients biochemically identified as having laboratory antiplatelet resistance are more likely to also have clinical resistance to antiplatelet agents.5–11 This is because they exhibit markedly higher risks of clinical adverse events than those who are identified as antiplatelet-sensitive.
Treatment of antiplatelet resistance
Increased dose of antiplatelet agents, change to other antiplatelet drugs and addition of more antiplatelet agent may be suitable for use in potential personalised strategies for antiplatelet resistance.12–18 The present guidelines about the necessity and recommendation of platelet function monitoring and tailored antiplatelet strategy are still uncertain, and the class of evidence is IIb (usefulness/efficacy less well established by evidence/opinion). The USA Food and Drug Administration box warning on clopidogrel in 2010 stated that the drug has diminished effectiveness in people with a poor ability to metabolise CYP2C19 and suggests alternative treatment strategies for these patients.20 A class IIb recommendation from the 2006 American College of Cardiology/American Heart Association guidelines stated 150 mg/d clopidogrel if <50% inhibition of platelet aggregation is demonstrated by platelet aggregation, as demonstrated in studies in patients undergoing PCI who are at risk of subacute stent thrombosis.21 The 2011 American College of Cardiology/American Heart Association guidelines stated that platelet function testing to determine platelet inhibitory response in patients with unstable angina/non-ST segment elevation myocardial infarction (UA/NSTEMI) (after ACS and PCI) on thienopyridine (IIb, B) and genotyping for a CYP2C19 loss of function variant in patients with UA/NSTEMI on clopidogrel might be considered if results of testing may alter management (IIb, C).22
Present findings and heterogeneity
In the present study, less risk of death or stent thrombosis was observed among antiplatelet-resistant patients who underwent personalised antiplatelet strategies than those who received conventional antiplatelet strategy, and there was no heterogeneity among these included studies. Moreover, tailored antiplatelet therapy in antiplatelet resistant patients was significantly associated with a decreased risk of total clinical adverse events, but with a statistical heterogeneity. The existence of statistical heterogeneity among studies can be partially explained by differences in clinical settings and methods of assessment, such as different countries, methods of determining platelet function and duration of follow-up according to the results of analysing subgroups. No additional risk of bleeding in tailored antiplatelet arm was compared with conventional antiplatelet arm in resistant patients. It seemed to indicate the potential safety of tailored antiplatelet therapy; of course, this was limited to the strategies used in the included studies here.
Other interesting studies
Some interesting trials related to our topic were excluded for not fulfilling the inclusive criteria. The RESPOND study and Alexopoulos et al observed that ticagrelor and prasugrel therapy could overcome clopidogrel non-responsiveness by demonstrating an increased inhibition of platelet aggregation in laboratory.23 ,24 However, no clinical outcomes of interest were reported. The CREST and MADONNA studies confirmed the benefit of treatment modification in antiplatelet-resistant patients, but in a non-randomised way.25 ,26 Recently, a multicentre, randomised trial organised by Collet et al27 showed no significant improvements in clinical adverse events with respect to platelet function monitoring and adjustment to treatment for coronary stenting after 1 year, as compared with standard antiplatelet therapy without monitoring. In contrast, the prospective randomised RAPID GENE trial reported treatment of identified CYP2C19*2 carriers with prasugrel can reduce HTPR.28 These two large studies were excluded from the present study because of different study design for randomisation. The structural differences between these two trials and the studies included are shown in online supplementary figure S5.
Study limitations
The results of this meta-analysis may have limitations. First, the number of studies and patients included were both relatively low. The topic—tailored antiplatelet therapy according to individual response—is a new and developing concept in clinical cardiology. Not many related studies have been published. Second, adverse clinical events were pooled from different trials even though they did not define these events in the same way. Third, there was significant heterogeneity with respect to the total number of adverse clinical events; this may be explained by a few potential effectors, such as those given above. Fourth, the total number of events remained relatively low despite pooling the data from 3738 antiplatelet-resistant patients, and thus there is no subgroup according to each different event, which limits the strength of conclusions regarding differences in clinical outcomes.
Key messages
-
What is already known on this subject
-
The results of studies evaluating the association between tailored therapy for antiplatelet resistance and decreased risk of adverse clinical events relative to conventional antiplatelet therapy have not been consistent.
-
What this study adds
-
For the first time, we performed a systematic review and meta-analysis of randomised trials to evaluate the clinical impact of tailored antiplatelet treatment relative to conventional antiplatelet treatment in antiplatelet-resistant patients.
-
How might this impact on clinical practice
-
In the study, personalised antiplatelet treatment for antiplatelet resistance was found to reduce the occurrence of death or stent thrombosis and total clinical adverse events when compared with the conventional treatment, and not to increase bleeding complications. It may provide important and useful evidence for clinical antiplatelet practice.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online figure-legends
- Data supplement 2 - Online figures
- Data supplement 3 - Online table 1
- Data supplement 4 - Online table 2
- Data supplement 5 - Online table 3
Footnotes
-
Contributors All authors have read and approved the final version of the manuscript. All authors have significantly contributed to the conception and design of the study, the analysis and interpretation of data, drafting of the manuscript or revising it to justify authorship.
-
Funding The National Natural Science Foundation (No.81100218).
-
Competing interests None.
-
Ethics approval Ethics review board varied according to each included study.
-
Provenance and peer review Not commissioned; externally peer reviewed.