Antiplatelet Therapy in the Era of Drug-Eluting Stents: Current and Future Perspectives
Gregory Ducrocq; Victor Serebruany; Jean-Francois Tanguay Expert Rev Cardiovasc Ther. 2007;5(5):939-953. ?2007 Future Drugs Ltd.
Abstract and Introduction
The use of drug-eluting stents (DESs) dramatically reduced in-stent restenosis. However, the increasing use of these stents has raised concern about their potential thrombogenicity. Indeed, the particularity of DES thrombosis compared with bare metal stent thrombosis is a high rate of late thrombosis. Antiplatelet therapy is efficient in preventing DES thrombosis. However, this therapy could be optimized and may be improved in the future. This article will review the mechanisms and the epidemiology of stent thrombosis. Then, we will summarize the antiplatelet therapeutic strategies used to prevent stent thrombosis and especially DES-associated thrombosis. Finally, we will present some data with regard to potential advantages and pitfalls in DES thrombosis prevention using novel antiplatelet agents currently under development, as well as future stent designs with improved healing properties.
Since the first angioplasty performed in 1977 by Gruntzig, improvements in techniques and devices have allowed a large diffusion of this revascularization procedure for coronary artery disease atherothrombosis and have also optimized the degree of reperfusion in ST-elevation myocardial infarction (STEMI). Intravascular metallic stents significantly reduced the risk of procedure-related complications but only provided a limited solution to the emerging problem of restenosis.[3-5] Drug-eluting stents (DESs) were introduced recently as an advanced alternative to bare metal stents. These hybrid devices are complex structures that associate a metallic stent, a polymer and various medications inhibiting intimal hyperplasia. Indeed, DESs dramatically reduce the rate of restenosis;[6,7] however, they may represent a trade-off with regard to the higher risks of stent thrombosis. This article will review the mechanisms and the epidemiology of stent thrombosis. Then, we will summarize the antiplatelet therapeutic strategies used to prevent stent thrombosis and especially DES-associated thrombosis. Finally, we will present some data with regard to potential advantages and pitfalls in DES thrombosis prevention of novel antiplatelet agents currently under development, and future stent designs with improved healing properties.
Pathophysiology of Stent Thrombosis
Under physiological conditions, endothelial cells play a pivotal role in the modulation of hemostasis, with a balance between secretion of prothrombotic (von Willebrand factor [vWF], tissue factor and plasminogen activator inhibitor) and antithrombotic molecules (thrombomodulin, tissue plasminogen activator, prostacyclin, nitric oxide and heparin sulphate).[8,9]
Balloon angioplasty and coronary stenting cause traumatism of the arterial wall, which is, in turn responsible for the loss of the endothelial cell barrier capacities, and which exposes subendothelial extracellular matrix (collagen) and lipid core components to the blood flow. The platelet's answer to this vascular injury can be divided into three steps that lead to thrombosis: adhesion, aggregation and activation.
Collagen and vWF are the main components of the extracellular matrix that are involved in the stimulation of platelet adhesion. Collagen binds to platelet glycoprotein α2β1 (GPIaIIa) and GPVI. However, under high shear stress conditions, a preliminary interaction between GPIbα (a leucine-rich subunit of the GPIb-IX-V complex) and vWF is necessary. Platelet activation can be turned on by this initial adhesion step and thereafter may cause important feedback loops implicating multiple agonists (Figure 1). Platelet activation leads to cell shape changes, secretion from α-granules and dense granules, and activation of the glycoprotein IIbIIIa (GPIIbIIIa) on the platelet surface. Activation of the GPIIbIIIa receptor allows fibrinogen binding; thus, platelet aggregation may occur and lead to thrombus extension. Finally, fibrin formation will then stabilize the platelet-rich thrombus.
Modern antiplatelet therapy can prevent or modulate many aspects and targets of this thrombotic cascade. However, in some cases, stent thrombosis may occur despite even aggressive antiplatelet treatment. As described by Honda et al, for bare metal stents (BMSs), physiopathologic factors involved in stent thrombosis can be divided into three groups and include intrinsic stent thrombogenicity, the patient himself and target lesions or procedure-related characteristics (Figure 2).
The increasing use of DESs has raised some concerns about their particular thrombogenicity. Clinical studies have identified new risk factors for DES thrombosis ( Table 1 ) and potentially new thrombotic mechanisms are being described for DES.
Some studies have suggested the potential specific thrombogenicity of the drug[26-28] or the polymer but the central point seems to be the impaired re-endothelialization. Indeed, animal studies have shown that DESs could delay endothelium formation and could even be responsible for incomplete re-endothelialization.[30,31] Moreover, Finn et al. have shown that overlapping DESs could delay even more endothelialization. Pathologic observations confirmed impaired endothelialization in man.[32,33] Joner et al. studied 23 autopsy cases of DESs matched with 25 cases of BMSs. There was a significant difference in the endothelialization score between DES (55.8%) and BMSs (89.8%). This delay in re-endothelialization was confirmed in vivo using angioscopy by Kotani et al. who studied 25 patients between 3 and 6 months after implantation of a sirolimus-eluting stent (SES) or a BMS. Three of the 15 SES (20%) had no neointimal coverage and only two (13.3%) had complete coverage. By contrast, all 22 BMSs demonstrated complete intimal coverage. This problem of impaired re-endothelialization may play a central role in some cases of late and very late DES thrombosis but may not totally explain this untoward clinical event.
Clinical studies have identified risk factors or independent predictors for DES thrombosis ( Table 1 ). Some of them were already known as risk factors for BMS thrombosis, including low ejection fraction, acute coronary syndrome at presentation,[17,18] residual reference segment stenosis, length, diameter and underexpansion of the stents. Others are only described as DES thrombosis risk factors, such as: renal failure, diabetes,[16,18,21] bifurcation lesions[16,24] or in-stent restenosis. However, head-to-head comparisons between BMSs and DESs are missing in these specific situations and these new risk factors may be related to specific populations included in DES studies. The most important risk factor for DES thrombosis is premature antiplatelet therapy discontinuation[16,17,19-21] (defined as a cessation of clopidogrel and/or aspirin before 3 months after implantation for a sirolimus-eluting stent and before 6 months for a paclitaxel-eluting stent). As described by Iakovou et al, stent thrombosis occured in 29% of patients with premature antiplatelet therapy discontinuation. Jeremias et al., Park et al, and Kuchulakanti et al., also described premature antiplatelet therapy discontinuation as a major risk factor for DES thrombosis (57 vs 1.7%, p < 0.001, Park et al.). While antiplatelet therapy discontinuation seems to remain a risk factor for DES thrombosis even after the delays described previously, it does not seem to be a significant risk factor for BMS thrombosis after 6 weeks.
Epidemiology of Stent Thrombosis
The analysis of stent thrombosis epidemiology is a challenging issue. indeed, in the early studies there was no consensus regarding the definition of stent thrombosis. Some studies only took into account angiographic-confirmed stent thrombosis. Regarding the very high mortality of stent thrombosis, other studies chose to include unexplained deaths in the definition of stent thrombosis. These differences in stent thrombosis definition may not allow direct comparisons in the incidences of stent thrombosis between different studies. More recently, in order to standardize the DES thrombosis definition, the Academic Research Consortium (ARC) proposed the following definitions. A 'confirmed thrombosis' was defined as an acute coronary syndrome (ACS) and an angiographic or pathologic confirmation of acute thrombus. A 'probable thrombosis' was defined as an unexplained death within 30 days of stent implantation or a target vessel myocardial infarction (MI) without angiographic confirmation of thrombosis or other identified culprit lesion. A 'possible thrombosis' was defined as any unexplained death beyond 30 days.
Bare Metal Stent Thrombosis
From the historic perspective, antithrombotic therapy was introduced as a combination of aspirin and a vitamin K antagonist in the initial BMS studies. With such a treatment, intra-hospital stent thrombosis occurred in 3.5 and 3.4% in the Belgium-Netherlands Stent (BENESTENT) and in STRESS studies, respectively. In these early stent studies, the mean delay for the occurrence of stent thrombosis was 6 days after angioplasty.
The introduction of thienopyridins, first ticlopidine and then clopidogrel, dramatically reduced the rate of acute and sub-acute stent thrombosis. In a meta-analysis, Cutlip pooled six studies in which the use of a thienopyridine was systematic and consistent (Stent Anticoagulation Restenosis Study [STARS], ACS multi-link Stent Clinical Equivalence in de Novo lesions Trial [ASCENT], Study of Microstent's Ability to limit Restenosis Trial [SMART], NIR Vascular Advanced North American Trial [NIRVANA], Evaluation of XT Stent for Restenosis in Native Coronary Arteries [EXTRA] and Coronary Stent System [CCS]) for a total of 6186 patients. The incidence of BMS thrombosis in this meta-analysis was 0.9%. By contrast, Moussa and Karrillon found an incidence of almost 2% in their studies.[13,15] The incidence of BMS thrombosis was found to be higher in the cases of multivessel angioplasty. In the Arterial Revascularization Therapies Study (ARTS), the incidence of stent thrombosis was 2.8% with a mean of 2.6 stented lesions per patient. However, the specific definition of stent thrombosis appears to be a key issue for a more refined interpretation of the incidence of this adverse event. For example, Cheneau et al. defined stent thrombosis as an angiography-proven stent occlusion less than 7 days after angioplasty. With such a limited definition, the incidence of stent thrombosis was only 0.4%.
In the initial randomized clinical trials, the incidence of DES thrombosis was less than 1% (0.4% at 1 year in the Sirolimus-Eluting Stent [SIRIUS] study[6,41] and 0.6% at 9 months in the TAXUS IV study). None of the randomized data found any statistically significant difference between the incidence of thrombosis associated with DESs or BMSs. However, in light of the low incidence of such an event, the randomized studies may have been underpowered to demonstrate such a difference. This is the main reason why attention has recently focussed on several pooled analyses. Camenzind et al. presented a pooled analysis of randomized clinical trials that found a trend toward more frequent death and Q wave MI for SESs (but not for paclitaxel-eluting stents) compared with BMSs. These results were consistent with a recent meta-analysis showing a significant increase in late thrombosis with DESs compared with BMSs. However, those data are conflicting with the meta-analysis of Spaulding, Kastrati or Mauri, which did not suggest any significant difference for the risk of stent thrombosis, as well as for the overall survival and survival free of MI at a 4 years follow-up between SESs and BMSs. In the Stone meta-analysis, an increase in late stent thrombosis for DESs after 1 year was observed but no significant difference in the cumulative rates of death or MI at 4 years has been established. Again, the conflicts reported above may be related to differences in stent thrombosis definitions. Furthermore, some studies censored stent thrombosis that occured after target-vessel revascularization, introducing a bias against DESs.
The relatively strict selection of patients reflected by the inclusion and exclusion criteria utilized in these randomized clinical trials made difficult any extrapolation to the 'real life' DES thrombosis clinical scenarios. Indeed, later randomized studies, which included less strict patient selection showed a higher rate of stent thrombosis (e.g., Sirolimus-Eluting Stent Compared with Paclitaxel-Eluting Stent for Coronary Revascularization [SIRTAX] study found an incidence of 2% in an unselected real-life cohort). Moreover, it was estimated that in the real world, as much as 60% of DESs were implanted for 'off-label' indications, or in types of lesions or patients that were excluded from the original pivotal studies that had led to their approval. Several large clinical registries indicated that the incidence of DES thrombosis in the contemporary percutaneous coronary intervention (PCI) world might be higher than the one reported from randomized trials. Iakovou et al. reported in their registry a 1.3% cumulative DES thrombosis at 9 months. Two institutional hospitals (Bern and Rotterdam) pooled their data in a large register including more than 8000 patients treated with a DES. The definition of stent thrombosis in this registry was rather restrictive since it only took into account angiographically proven cases of DES thrombosis. The total incidence rate of DES thrombosis was 2.9% and, more interestingly, late stent thrombosis occurred at a constant rate of 0.6% per year for up to 3 years. This higher risk in late DES stent thrombosis could result in severe clinical events, as suggested in the basel stent cost-effectiveness trial late thrombotic events BASKET LATE trial that showed an increase in late cardiac death or nonfatal MI, possibly related to late stent thrombosis. In a large Swedish registry including nearly 20,000 nonrandomized patients, the data suggested early benefit with DESs that reversed later to an increase in the risk-adjusted rate of death for DESs compared with BMS (adjusted relative risk of 1.18 at 3 years). However, those nonrandomized comparisons have important potential limitations (selection biases, limited follow-up) that preclude any strong conclusion.
Although conflicting, these emerging clinical data have raised legitimate concerns and led to a special US FDA meeting in December 2006 to review the evidence of these safety concerns of DESs and the use of dual antiplatelet therapy with different stakeholders from the industry, the medical and scientific community, and the regulatory agency. An important conclusion was that when DESs are used for their approved indications, the risk of thrombosis does not outweigh their advantages over BMSs in reducing the rate of repeated revascularization. However, the panel also concluded that, compared with on-label use, off-label use is associated with increased risks of both early and late DES thrombosis, as well as death and MI.
Timing of Stent Thrombosis
In the Cutlip et al. meta-analysis, most of the BMS thrombosis occurred during the first 48 h after intervention. The rate of late BMS thrombosis is not well known because most clinical studies had a follow-up limited to 30-180 days; however, it is believed to be very low. The control arms of more recent trials with DESs strongly suggest a very low incidence of late BMS thrombosis.
The particularity of DES thrombosis is the frequency of late and very late thrombosis. For example, in the Iakovou study, 52% of DES thrombosis occurred more than 30 days after angioplasty. This has led to the definition for DES thrombosis according to its timing: acute thrombosis occurs less than 24 h after angioplasty; subacute thrombosis between 24 h and 30 days; late thrombosis between 30 days and 1 year and very late thrombosis after 1 year. Interestingly, in the Bern-Rotterdam registry, late stent thrombosis occurred at a constant rate of 0.6% per year for up to the 3 years reported. Whether this rate will remain constant in the following years remains to be determined by longer term follow-up.
Prognosis of Stent Thrombosis
In the Cutlip et al. meta-analysis, the 6-month mortality rate after a BMS thrombosis was 20%, whereas it was only 1% in the group that did not suffer stent thrombosis. Those results are similar to the 30-day mortality rate reported by Moussa et al. (26%) and Karrillon et al. (24%) after BMS thrombosis. Interestingly, the death rate for DES thrombosis has often been reported to be higher than the mortality rate for BMS thrombosis. Iakovou et al. observed a mortality rate of 45% at 9 months and Ong et al. reported a 29% incidence of death after late DES thrombosis. The exact reason for this possible difference in mortality is unclear but could be related to the mechanism of action of DESs and the impact of drug release on collateral functions.
Stent & Antiplatelet Therapy
As discussed previously, the most important risk factors for DES thrombosis is premature discontinuation of antiplatelet therapy and/or noncompliance. Therefore, it may be very important to review the guidelines regarding antiplatelet therapy and improve education to correct any potential gap in the knowledge. Is there a possibility to improve the antiplatelet protection by increasing the dose or the duration of the antiplatelet, therapy? Conversely, a significant number of DES thrombosis occurred while under dual antiplatelet therapy. For example, in the Bern-Rotterdam Registry, 87% of patients reported with early stent thrombosis were on dual antiplatelet therapy. By contrast, late stent thrombosis occurred during dual antiplatelet therapy in only 23% of cases and while on single antiplatelet drug therapy in 51% of cases. Could more powerful antiplatelet drugs avoid some of the early or late stent thrombosis? What are the newer antiplatelet drugs that could offer a potential alternative or additional therapy in the future?
The mode of action of aspirin as an antiplatelet agent is through an irreversible acetylation of the serine 509 of the COX-1 enzyme resulting in the inhibition of thromboxane A2 generation in the platelet. Reduction in the production of this potent platelet agonist can significantly reduce platelet aggregation.
Aspirin was the first antithrombotic treatment used after stenting. The initial studies using this single antiplatelet therapy reported a very high rate of stent thrombosis, ranging from 15 to 20%.[2,53,54]
However, despite all the shortcomings, aspirin remains the cornerstone of antiplatelet therapy after stenting. The American College of Cardiology/American Heart Association/ Society for Cardiac Angiography and Interventions Task Force of Practical Guidelines (ACC/AHA/SCAI) 2005 guidelines for PCI recommended that patients already taking daily chronic aspirin therapy should take 75-325 mg of aspirin before the PCI procedure is performed (Class I Level of Evidence: A). Patients not already taking daily chronic aspirin therapy should be administered 300-325 mg of aspirin at least 2 h, and preferably 24 h, before the PCI procedure is performed. (Class I Level of Evidence: C). Higher doses of aspirin are recommended for patients not already taking aspirin therapy immediately before PCI procedures.
After the PCI procedure, in patients with no known allergy or increased risk of bleeding, aspirin 325 mg daily should be administered for at least 1 month after BMS implantation, for 3 months after SES implantation and finally, for 6 months after paclitaxel-eluting stent implantation; after which chronic aspirin therapy following stenting should be continued indefinitely at a daily dose of 75-162 mg (Class I Level of Evidence: B).
Recent European guidelines recommended for patients not chronically pretreated, that a loading dose of 500 mg orally should be administered more than 3 h prior to, or at least 300 mg intravenously directly prior to, the procedure. Moreover, after angioplasty there is no need for doses higher that 100 mg daily (Class I Level of evidence: B).
Although several large randomized studies showed a superiority of the addition of dipyridamole over aspirin for second ischemic stroke prevention, such a benefit has never been shown after coronary angioplasty. In a randomized study of 232 patients, Lembo et al. did not demonstrate any benefit with the addition of dipyridamole to aspirin before elective percutaneous transluminal coronary angioplasty (PTCA) compared with aspirin alone. However, this study was conducted in the pre-stent era and with a small sample size. A larger study evaluating the addition of dipyridamole over aspirin and clopidogrel after stenting would be useful.
Vitamin K Antagonists
In the initial BMS studies, the association of vitamin K antagonists to aspirin translated to a reduction in the rate of stent thrombosis to as low as 4%.[3,4] However, this association led to a very high risk of hemorrhagic complications. For example, in the BENESTENT study, bleeding complications were 13.5% in the stent group receiving such treatment versus 3.1% in the balloon angioplasty group receiving aspirin alone.
Mechanism of action. Although ADP is a weak platelet agonist, it plays a key role because it is secreted by platelets itself from dense granules where it is stored. By this mechanism, it amplifies the platelet answer induced by other agonists. ADP has two platelet receptors: the P2Y1 receptor leads to platelet shape change and rapidly reversible aggregation; the P2Y12 leads to a sustained aggregation. Activation of both receptors is necessary for normal implementation of ADP-induced platelet aggregation.
Thienopyridines are P2Y12 antagonists. There are two thienopyridines currently used in clinical practice: ticlopidine and clopidogrel. Both are prodrugs that need to be metabolized by P-450 cytochrome in order to be tranformed into active metabolites. The mechanism of action of thienopyridines is an irreversible inhibition of the P2Y12 platelet receptor.
Clinical studies. Dual antiplatelet therapy with aspirin and ticlopidine significantly reduced cardiac events after implantation of a BMS compared with aspirin alone or aspirin and vitamin K antagonists.[37,59-61] This protecting effect resulted mainly from a reduction of acute and subacute stent thrombosis with an incidence of less than 1%. Moreover, most studies showed a dramatic reduction in hemorragic complications compared to combined therapy with aspirin and vitamin K antagonists. In the Full Anticoagulation versus Aspirin and ticlopidine (FANTASTIC) study hemorrhagic events were 13.5% for the aspirin and ticlopidine arm versus 21% in the aspirin and vitamin K antagonist arm. Despite its superiority, the use of ticlopidine was limited. The hematologic toxicity of this compound led to its substitution by clopidogrel. Three randomized studies[63-67] and a meta-analysis showed that clopidogrel was at least as effective as ticlopidine, well tolerated and a much safer drug.
Interest of a loading dose & of a prolonged therapy. The Clopidogrel for Reduction of Events During Observation (CREDO) trial was a randomized trial conducted in 2116 patients undergoing elective PCI with a BMS. Patients were randomly assigned to receive a 300 mg loading dose of clopidogrel or no loading dose before PCI. Thereafter, all patients received clopidogrel 75 mg daily until day 28. From day 29 to 12 months, patients in the loading dose group continued receiving clopidogrel 75 mg/day while those in the control group received a placebo. The 12-month incidence of the composite of death, MI or stroke was reduced by 26.9% in the group with a clopidogrel loading dose and a long-term treatment. The PCI- Clopidogrel in Unstable angina to prevent Recurrent Events (CURE) trial evaluating 2658 patients with non-stent thrombosis elevation ACS enrolled in the randomized CURE study and treated by angioplasty (BMS population) showed a 31% reduction in cardiovascular events in the long-term clopidogrel group. These two studies justify the interest of a sustained dual antiplatelet therapy, as well as using a 300 mg clopidogrel loading dose.
However, the timing of this loading dose before angioplasty appears to be important. In the CREDO trial, the 300 mg clopidogrel loading dose seems to only be beneficial if administered at least 6 h before angioplasty and when followed by clopidogrel 75 mg daily. Moreover, a recent post hoc analysis of this trial showed that the difference in outcomes at 28 days between placebo and clopidogrel pretreated patients was not significant if the 300 mg loading dose was administered less than 15 h before angioplasty, underlying the delayed efficacy of the 300 mg loading dose. In PCI-CURE, which included only patients with ACS without stent thrombosis elevation, administration of a 300 mg loading dose as soon as possible before angioplasty followed by 75 mg clopidogrel for 9-12 months has been associated with a significant reduction of death/MI.
Guidelines. ACC/AHA/SCAI 2005 guidelines recommend that patients should be loaded with clopidogrel before PCI is performed (Class I Level of Evidence: A). An oral loading dose of 300 mg, administered at least 6 h before the procedure, has the best established evidence of efficacy (Class I Level of Evidence: B). In patients who have undergone PCI, clopidogrel 75 mg daily should be given for at least 1 month and could be continued for 1 year after BMS implantation (unless the patient is at increased risk for bleeding, then it should be given for a minimum of 2 weeks; Class I Level of Evidence: B). When a loading dose of clopidogrel is administered, a regimen of greater than 300 mg is reasonable to achieve higher levels of antiplatelet activity more rapidly, but the efficacy and safety compared with a 300-mg loading dose are less established (Class I Level of Evidence: C).
ESC guidelines recommend administering a loading dose of 300 mg clopidogrel at least 6 h before angioplasty in stable angina (grade IC) and as soon as possible in ACS (grade IB), followed by an association of aspirin and clopidogrel during 3-4 weeks following angioplasty (grade IA) in stable angina and 9-12 months in the case of ACS after implantation of a BMS.
The American College of Chest Physicians ACCP guidelines recommend administration of aspirin and clopidogrel for at least 1 month following stent implantation and then to administer aspirin indefinitely (grade IA).
These guidelines are for BMSs. For DESs, owing to the risk of late thrombosis, randomized trials have maintained clopidogrel for a longer time (3 months for Cypher?, 6 months for Taxus?). ACC/AHA/SCAI guidelines recommend dual therapy for 3 months after sirolimus stent implantation, and 6 months after paclitaxel stent implantation, and ideally up to 12 months in patients who are not at high risk of bleeding (Class I Level of Evidence: B).
European guidelines (ESC 2005) recommend continuing dual antiplatelet therapy (aspirin and clopidogrel) for 6-12 months after angioplasty (grade IC). No specific recommendation was made for loading dose before DES implantation.
DES & thienopyridine: daily practice. In the PREMIER registry, among 500 DES-treated STEMI patients, 13.6% stopped thienopyridine therapy within 30 days. The only factor independently associated with clopidrogrel discontinuation was low education level (less likely to have completed high school). The mortality rate was higher in the group who discontinued thienopyridine therapy (7.5 vs 0.7%). These results emphasize the need for an adapted patient education strategy before hospital discharge.
Grines et al. addressed several practical recommendations to prevent premature thienopyridine discontinuation and to reduce the risk of DES thrombosis, including :
The Canadian Association of Interventional Cardiology (CAIC) and the Canadian Cardiovascular Society (CCS) summarized the available evidence and provided practical recommendations regarding DES use and antiplatelet therapy. Several statement points are that physicians should always carefully consider the benefits and risks on an individual patient basis when choosing between a DES and a BMS. In addition, in this selection process, the interventional cardiologists should be even more careful when considering DES use for unapproved (off-label) indications and be meticulous in deployment techniques. As for the duration of dual antiplatelet therapy, even longer term than 1 year could be considered in patients treated with DES who are thought to be at higher risk for very late stent thrombosis or in whom such an event is likely to have fatal consequences. However, the exact duration of treatment should be determined on an individual basis after careful consideration of the competing risks of stent thrombosis and bleeding. There was no evidence to support restarting clopidogrel in patients who have completed their course of dual antiplatelet therapy and remained event-free on aspirin monotherapy.
Finally, the Society for Cardiovascular Angiography and Interventions Drug-eluting Task Force published an excellent report with very practical clinical advices. This clinical alert document emphasizes the medical-legal aspects, including the importance of close communication between the physician and the patient with careful documentation of the decision process in the medical record. In addition, the report highlights the crucial need for more studies and encourages recruitment in several large multicenter trials (Harmonizing Outcomes with Revascularization and Stents in Acute MI [HORIZONS-AMI], International Nifedipine GITS study: Intervention as a Goal in Hypertension Treatment [INSIGHT], Patient Related OuTcomes with Endeavor versus Cypher stenting Trial [PROTECT], Future Revascularization Evaluation in Patients with Diabetes Mellitus: Optimal Management of Multivessel Disease [FREEDOM] and Synergy between Percutaneous Coronary Intervention with Taxus and Cardiac Surgery [SYNTAX]).
How long should we maintain dual antiplatelet therapy in patients treated with a DES? In 2004, Mc Fadden et al. first described four cases of angiographically-documented DES thromboses that occurred between 343 and 442 days after implantation. All cases arose soon after antiplatelet therapy was interrupted. This clinical concern regarding late and very late DES thrombosis after antiplatelet therapy cessation was recently confirmed by Pfisterer et al. in a substudy of BASKET-LATE trial, which found that after clopidogrel discontinuation at 6 months, documented late stent thrombosis and related death/target vessel MI, between 6 and 18 months after angioplasty, were twice as frequent after DESs versus BMS (2.6 vs 1.3%). These results raise questions regarding the optimal duration of thienopyridine therapy. Should thienopyridine therapy be pursued beyond recommendations (i.e., beyond 12 months)? If so, should we do this for all patients treated with a DESs or should we select only 'high-risk patients'? Even if there is a strong temptation to answer affirmatively to the first question regarding the data reported by Pfisterer et al., we must remain cautious and use our clinical judgment. Prolonged antiplatelet therapy may increase the bleeding risk and such a decision of long-term dual antiplatelet therapy must be balanced.
Currently, we still have no definite data to answer the key question of usefulness of prolonged thienopyridine therapy in patients treated with DESs beyond 1 year. We are eagerly awaiting longer term follow-up, large prospective randomized trials and more data on the use of DESs and antiplatelet therapies from large registries. In addition, a randomized clinical study comparing clopidogrel and placebo after 1 year post-PCI would be very helpful to definitely answer such a question with current DESs.
Is the recommended posology of clopidogrel optimal? Some studies suggest that there is a great variability of platelet response after a standard loading dose of 300 mg of clopidogrel.[77-79] Others suggest that a loading dose of 600 mg would have a more rapid and more powerful antiplatelet effect.[80-82] The antiplatelet effect of a 600 mg loading dose seems to be maximal after 2-3 h.[80-85] Moreover, such a loading dose would be able to decrease the percentage of 'low clopidogrel responder' patients.
Clinical studies have confirmed the platelet data described previously. In a retrospective analysis of the TARGET trial, pretreatment with a 300 mg clopidogrel loading dose significantly decreased the risk of death and MI at 30 days and 6 months and the risk of death at 1 year; with a stronger difference when the loading dose was given at least 6 h before angioplasty. The Intracoronary Stenting and Antithrombotic Regimen - Rapid Early Action for Coronary Treatment (ISAR-REACT) trial strongly suggested a protective effect of a 600 mg loading dose. In this study, including low-risk patients receiving a 600 mg clopidogrel loading dose at least 2 h before angioplasty, abciximab did not show any clinical benefit against placebo. The 'Intracoronary Stenting and Antithrombotic Regimen - Is Abciximab a Superior Way to Eliminate Elevated Thrombotic Risk in Diabetics?' (ISAR-SWEET) trial, showed similar results in a population of diabetic patients. However, no comparison between a 300 and a 600 mg clopidogrel loading dose was performed in these two trials.
The ARMYDA-2 trial randomized 255 patients between a 300 mg clopidogrel loading dose and a 600 mg loading dose 4 to 8 h before angioplasty. The 600 mg group showed a significative decrease in cardiovascular events (4 vs 12%) without increasing the hemorrhagic risk. Similarly, Cuisset et al. randomized a total of 292 consecutive NSTE ACS patients undergoing coronary stenting between a 300 and a 600 mg loading dose of clopidogrel at least 12 h before percutaneous coronary intervention. The ADP-induced platelet aggregation and expression of P-selectin were significantly lower in patients receiving 600 mg of clopidogrel compared with those receiving only 300 mg, which translated into a significant reduction in cardiovascular events at 1 month in the 600 mg group (5 vs 12%).
Future Developments to Reduce DES Thrombosis
Platelet Function Tests
Variability of clopidogrel response When assessed with biological tests, the percentage of patients described as 'resistant' or 'low responders' to clopidogrel varies between 4 and 30% in the literature. These important variations are linked to a lack of consensus regarding the definition, laboratory technics and the threshold that should be used to identify those low responder patients.
Gurbel et al. assessed platelet function with light transmittance aggregometry after stimulation by ADP and defined 'resistance' as less than 10% reduction of aggregation between baseline and a time point after clopidogrel loading. With these methods and definitions, 'resistant patients' were 31% at 5 days and 15% at 30 days.
Serebruany et al. also studied platelet function with light transmittance aggregometry after stimulation by ADP before and after a clopidogrel loading dose. The distribution of responsiveness follows a normal curve (Figure 3). Low and high responders were defined as those varying more than 2 standard deviations to the mean. With this definition, low responders were only 4.2%.
Potential consequences of clopidogrel response variability Are clopidogrel low responders or resistant patients more likely to have stent thrombosis? Four retrospective studies tried to answer this question:
These results support the hypothesis that patients who have higher platelet reactivity despite clopidogrel therapy (also called clopidogrel low responders or 'clopidogrel resistant') are more likely to suffer stent thrombosis. However, these four studies are retrospective assessments in which blood tests were performed early after stent thrombosis. It is difficult to know whether platelet activity is a cause or a consequence of stent thrombosis. However, a small prospective study performed in 105 patients undergoing elective angioplasty found that two patients developed stent thrombosis; both of which were clopidogrel low responders. Other prospective trials have shown a relation between MACE after angioplasty and lower response to clopidogrel.[98,99] More recently, a monocentric prospective study showed, for the first time, that clopidogrel nonresponsiveness (assessed by light transmitance aggregometry) was a predictor of stent thrombosis (HR: 3.08). However, larger multicentric prospective trials are important in order to collect stronger evidence that low clopidogrel responders are most likely to suffer stent thrombosis. Moreover, we need to elaborate standardized definition, techniques of evaluation and a standardized threshold of low clopidogrel response for a given technique. The optimal treatment in patients identified as clopidogrel low reponders is still to be defined: should we simply increase the clopidogrel dose or switch to an alternative antiplatelet regimen? In short, there are a lot of issues that urgently need to be resolved before the introduction of a possible individualized antiplatelet therapy based on serial platelet function tests.
As a matter of fact, there are no specific recommendations regarding the necessity of systematic evaluation of platelet function before angioplasty. However, ACC/AHA 2005 guidelines suggest that 'in patients in whom subacute thrombosis may be catastrophic or lethal (unprotected left main, bifurcating left main or last patent coronary vessel), platelet aggregation studies may be considered and the dose of clopidogrel incresed to 150 mg/day if less than 50% inhibition of platelet inhibition aggregation is demonstrated' (Level of Evidence: IIb C).
New DESs. New DESs are under development and the safety of these devices has become of major clinical importance.
The Endeavor? DES developed by Medtronic is a stent eluting zotarolimus (a drug from the same family as the sirolimus eluted from the Cypher stent) from a phosphorylcholine polymer. Its more rapid eluting profile could be critical for a better late endothelialization. Indeed, in the Endeavor II study, randomizing 1197 patients to the Endeavor stent versus a BMS, the incidence of stent thrombosis was significantly lower in the Endeavor group (0.5 vs 1.2%). Moreover, there was no late stent thrombosis in the Endeavor group up to 24 months. The 2-year follow-up of the Endeavor III study presented at the recent ACC Congress in 2007 appears to confirm these data with no late stent thrombosis up to 2 years in the Endeavor group. However, we need more data from registries to confirm if this DES is really safer in the real world while maintaining clinical efficacy. The PROTECT study that will randomize 8000 patients to Endeavor versus Cypher DESs with a 3-year follow-up and with stent thrombosis and mortality as the primary end points will also provide us with the currently lacking information regarding the safety of this device.
Bioresorbable materials. New concepts are emerging in order to increase the safety of DESs. Of the most interesting are bioresorbable polymers and bioresorbable stents.
The CoStar? stent is a device developed by Conor that contains small reservoirs eluting drugs from a bioresorbable polymer on the luminal or vascular surface. The initial small studies evaluating this device did not report any late stent thrombosis. However, the COSTAR II trial (EuroPCR 2007) showed a significantly higher MACE rate for the CoStar stent compared with the Taxus stent. This has interrupted the development of this stent. However, a new stent, based on the same platform, but eluting sirolimus, is under development.
Bioresorbable materials for the stent itself are currently under development. To achieve such a goal, magnesium could be the metal of choice. Poly-l-lactic acid could be even more promising, as shown in a porcine model. Patrick Serruys presented at the ACC congress in 2007 the first in-man results of a fully bioabsorbable DES made of poly-l-lactic acid and everolimus. This small feasibility study (only 30 nonrandomized patients included) could not reach any conclusions about the efficacy or the safety of this novel DES. While the concept of such bioabsorbable DESs is promising, larger studies are necessary to evaluate the future clinical utility of these devices.
Other strategies. 'Prohealing approaches' are also of interest to reduce stent thrombosis. A preliminary study in humans assessed the implantation of stents coated with CD34 antibodies in order to 'capture' endothelial progenitor cells. This strategy was shown to be both feasible and safe; however, the efficacy still needs to be determined.
Coating stents with antiplatelet drugs could also be a promising way to prevent stent thrombosis. A recent study randomized 96 patients in an acute STEMI context to an abciximab-coated stent or a BMS. During 1 year follow-up, two patients in the BMS group had a recurrent acute MI (AMI) whereas no patient in abciximab-coated stent suffered an AMI. Although this strategy seems intriguing, larger studies are required to draw appropriate conclusions.
Emerging Antiplatelet Therapy
Cilostazol. Cilostazol selectively inhibits 5'3'-cyclic nucleotide phosphodiesterase III and has antiplatelet and vasodilating effects. This agent also inhibits smooth muscle cell proliferation in vitro.
Dual Therapy. A meta-analysis pooled in 2004 studies comparing aspirin and ticlopidine with aspirin and cilostazol after angioplasty with BMS. There was no statistically significant difference between the two stategies for cardiac events. However, this comparison was limited by the small study size. In a study that randomized the use of a paclitaxel-eluting stent or a BMS, five out of 37 patients treated with cilostazol experienced stent thrombosis in comparison to one out of 138 patients treated with a thienopyridine. The antiplatelet therapy was not randomized in this study, but this finding raises concern about the safety and efficacy of cilostazol in patients treated with DESs.
Triple Therapy. In a single center nonrandomized study of more than 3000 patients, a Korean team compared a dual antiplatelet therapy (aspirin with clopidogrel or ticlopidine) to a triple antiplatelet therapy (aspirin with clopidogrel or ticlopidine with cilostazol) after implantation of a BMS. Stent thrombosis within 30 days was significantly lower with triple antiplatelet therapy than with dual antiplatelet therapy despite more complex lesions in the triple therapy group (longer stents, multivessels) and without a significant increase in bleeding complications.
This study has several limitations (nonrandomized, relatively small sample size, including only BMS) and no definite conclusion can be extrapolated. However, it could be worth evaluating a triple antiplatelet therapy in higher risk patients, in a randomized fashion, and with a much larger sample size.
P2Y12 inhibitors. Cangrelor Cangrelor is an ATP analogue, and an intravenous reversible P2Y12 inhibitor. The short plasma half-life (3-5 min) results in a full recovery of platelet activity within 60 min. Cangrelor is a strong platelet inhibitor and results in a near complete inhibition of platelet aggregation. Ongoing Phase III studies will bring us further information regarding the clinical efficacy and safety of this drug. However, cangrelor can only be administered intravenously, which will limit its utilization to the acute clinical conditions.
Prasugrel. Prasugrel is an oral thienopyridine. Its antiplatelet action is due to irreversible and selective blockade of P2Y12 by its active metabolite. It exhibits more potent and rapid platelet inhibition than clopidogrel in vitro. Phase I studies in healthy human subjects confirmed faster and greater inhibition of platelet aggregation than those achieved with clopidogrel. In addition, the prevalence of low responders seems to be lower with prasugrel than with clopidogrel.[114,115] These differences in platelet responsiveness may be due, in great part, to a more efficient generation of the active metabolite of prasugrel. Indeed, in vitro, the active metabolites of the two drugs were demonstrated to have similar potency.
Joint Utilisation of Medications to Block Platelets Optimally - Thrombolysis In Myocardial Infarction (JUMBO-TIMI 26), a Phase II safety clinical study, randomized 904 patients between clopidogrel and three regimens of prasugrel. Both clopidogrel and prasugrel resulted in low rates of bleeding.
Trial to assess Improvement in Therapeutic Outcomes by optimizing platelet Inhibition with prasugrel Thrombolysis In Myocardial Infarction (TRITON-TIMI-38) trial is the first Phase III study of prasugrel. It is ongoing and will randomize 13,000 ACS patients to prasugrel or clopidogrel for up to 15 months.
AZD 6140. AZD 6140 does not need metabolization to be active; however, it is metabolized into an active metabolite that participates to its activity. It is a reversible antagonist of P2Y12 receptor.
In a Phase II, double-blind trial, Husted et al. randomly assigned 200 atherosclerotic patients to receive AZD6140 50, 100 or 200 mg twice daily or 400 mg daily or clopidogrel 75 mg daily for 28 days. AZD6140 (100 and 200 mg twice daily, 400 mg daily) rapidly and nearly completely inhibited ADP-induced platelet aggregation at day 1 and day 28. After 4 h of the dose being administered at steady state, the three higher doses of AZD6140 produced a comparable final-extent mean percentage of inhibition of platelet aggregation (approximately 90-95%), which exceeded that with AZD6140 50 mg twice daily or clopidogrel 75 mg daily (approximately 60%).
A study of Platelet inhibition and patient Outcomes AZD (PLATO) is a Phase III trial that will enroll 18,000 patient suffering from ACS for randomization between AZD 6140 and clopidogrel. The primary end point will be a composite of cardiovascular events.
Platelet adhesion antagonists. Theoretically, all of the steps involved in platelet activation and coagulation are potential targets to prevent DES thrombosis after stent implantation. Blocking platelet adhesion to subendothelium (which is the first step of stent thrombosis) could be of particular interest.
Since the interaction between GPIbα is the very first step of platelet adhesion to subendothelium, blocking the GPIb-IX-V/vWF pathway could be advantageous. This strategy was shown to be effective in different animal models.[119-126]
Blocking GP VI/collagen interaction could be another strategy of platelet adhesion inhibition. This target of therapy was shown to prevent thrombosis in rodent models.[127-129] However, the efficacy of this strategy in the context of stenting has yet to be determined.
Stent thrombosis is a rare, but serious, complication of modern percutaneous interventions, which could lead to MI and death. The extensive use of DESs has raised concern about late and very late thrombosis. Antiplatelet therapy can be the cornerstone in reducing the incidence of DES thrombosis. Aspirin remains an indispensable therapy for every patient treated with a DES and should be pursued indefinitely.
Thienopyridines are a class that has also shown efficacy. It should be associated to aspirin for at least 1 year after a DES implantation. A better patient education and compliance could avoid premature antiplatelet drug cessation, which can be associated with catastrophic consequences. The two approved molecules (ticlopidine and clopidogrel) have response variability that may have clinical implications. The interest of identification of low responders with biologic tests is still to be determined. Higher doses of clopidogrel may reduce the rate of low responders, although the clinical significance of this phenomenon is not clear. New thienopyridins may also reduce the rate of low responders by a higher platelet inhibition. Some of these novel molecules may also be of great interest for the reversibility of their effects, which could become the therapy of choice before surgery in DES patients. While other antiplatelet therapies are under development, a triple association could achieve greater platelet inhibition.
The particularity of DES thrombosis is the higher rate of late and very late stent thrombosis. The main risk factor of DES thrombosis is premature antiplatelet therapy cessation. A thorough interrogating prior to catheterization is mandatory to avoid implanting DESs in patients not expected to comply with prolonged antiplatelet therapy or in patients who are likely to require invasive or surgical procedures within the following months. Before discharge, a better patient education about the importance of dual antiplatelet therapy and the consequences of stopping it prematurely should be given with documentation in the medical record.
Aspirin, the cornerstone of antiplatelet therapy after a DES implantation, must be maintained indefinitely. Clopidogrel should be used in combination with aspirin for at least 12 months in such patients who are not at high risk for bleeding. Even longer term dual antiplatelet therapy could be considered in patients treated with DES who are at higher risk for very late thrombosis or in whom stent thrombosis is likely to have fatal consequences.
While a significant number of DES thrombosis occurs while under dual antiplatelet therapy, the identification of low responders to antiplatelet therapy could help reduce these events with more powerful antiplatelet drugs currently under development. Phase III trials with promising new thienopyridines are currently ongoing and could reduce the rate and the clinical impact of the low responders patients. Novel classes of antiplatelet therapy, with targets such as the GPIb-IX-V/vWF or the GP VI/ collagen pathways, are emerging. These antagonists could also play a role in optimizing further prevention of DES thrombosis. Finally, the interest of a triple antiplatelet therapy (with cilostazol or another molecule) in patients with a very high risk of thrombosis will need to be evaluated.
Table 1. Clinically Identified Risk Factors of Bare Metal Stent and Drug-Eluting Stent Thrombosis
Sidebar: Key Issues
Gregory Ducrocq is a recipient of a research grant from the F??ation Francaise de Cardiologie. Jean-Francois Tanguay is from the Department of Medicine, University of Montreal (Quebec, Canada) and is supported in research by the Fonds de la recherche en sant?du Qu?ec (FRSQ), the Heart and Stroke Foundation of Canada (HSF), the Canadian Institutes of Health Research (CIHR). He also receives industry grants from Abbott, Astra-Zeneca, Boehringer-Ingelheim, Bristol-Myers-Squibb, Eli Lilly, Estracure, Pfizer and Sanofi-Aventis.
Jean-Francois Tanguay. Email: Jeanfirstname.lastname@example.org
Gregory Ducrocq, Universit?de Montr?l, Institut de Cardiologie de Montr?l, Research Centre, Montreal Heart Institute, Department of Medicine, 5000, rue B?anger, Montr?l, Qc H1T 1C8, Canada
Victor Serebruany, Johns Hopkins University, Department of Medicine, 9101 East Stayman Drive, Ellicott City Towson, MD 21204 USA
Jean-Francois Tanguay, Universit?de Montr?l, Institut de Cardiologie de Montr?l, Research Centre, Montreal Heart Institute, Department of Medicine, 5000, rue B?anger, Montr?l, Qc H1T 1C8, Canada.