ANMCO/SICI-GISE paper on antiplatelet therapy in acute coronary syndrome^†

Abstract

Introduction

Antiplatelet therapy is the cornerstone of the pharmacologic management of patients with acute coronary syndrome (ACS)¹ (Figure 1).

Figure 1

Possible mechanisms of platelet activation and mode of action of the main antiplatelet drugs.

Open in new tabDownload slide

The possibility to obtain optimal platelet inhibition was first raised by the advantages shown by aspirin (ASA) in cerebrovascular disease,^2,3 and then confirmed by the addition of ASA to a thienopyridine for the prevention of stent thrombosis.^4–7 Thienopyridines are pro-drugs whose active metabolites act by antagonizing ADP platelet receptor, resulting in the inhibition of platelet aggregation and activation.⁸ The association of ASA with ticlopidine, a first-generation thienopyridine, has shown a higher efficacy in reducing the risk for stent thrombosis when compared with ASA alone or to the association of ASA and warfarin.^4–7

In the clinical practice, clopidogrel—a second-generation thienopyridine—has largely replaced ticlopidine due to its larger bioavailability and convenience of use, as well as to its lower association with adverse events (mainly neutropenia).⁸ Clopidogrel inhibits platelet aggregation by selectively and irreversibly antagonizing subtype P2Y₁₂ of ADP platelet receptors; the active metabolite is produced following cytochrome P450-mediated biotransformation in the liver.⁸ The efficacy of long-term treatment with clopidogrel in addition to the standard treatment with ASA was confirmed by the Clopidogrel in Unstable angina to prevent Recurrent Events (CURE)⁹ trial, which demonstrated a significant reduction in the composite endpoint of cardiovascular mortality, non-fatal myocardial infarction and stroke in non-ST elevation ACS (NSTE-ACS) patients when compared with placebo. In this trial, 12 562 patients were randomized to a double-blind treatment with either placebo or a loading dose of 300 mg clopidogrel, followed by a daily dose of 75 mg for the subsequent 12 months.⁹ The administration of clopidogrel produced an early effect that was seen since the first day of treatment and was maintained for 1 year, with a 20% reduction in the relative risk (RR) of developing the composite endpoint.¹⁰ Each of the events included in the primary endpoint showed a similar pattern: taken individually, the risk of myocardial infarction was reduced by 23%, cardiovascular mortality by 7% and stroke by 14%. This benefit was accompanied by a significant increase in the risk of minor bleeding (whose incidence was approximately twice as that observed in the clopidogrel group) and major bleeding (RR increased by 30%), but not in the risk of life-threatening bleeding, with a net clinical benefit in favour of clopidogrel.^11,12 About one-fifth of patients in the CURE trial underwent percutaneous coronary intervention (PCI); this subset of patients was analysed in a perspective ad hoc study, PCI-CURE,¹³ whose objective was to determine whether the addition of clopidogrel to ASA prior to PCI, followed by a long-term therapy, reduced the incidence of the composite endpoint—which in this case included cardiovascular death, myocardial infarction, and urgent revascularization—when compared with ASA alone. The results showed a significant reduction of 30% in the endpoint for the group randomized to clopidogrel, with no significant differences in the rate of major bleeding.¹³ Both studies (CURE and PCI-CURE) validated the 300-mg dose of clopidogrel, while subsequent studies tested higher loading doses (600–900 mg) and showed a more rapid and intensive platelet inhibition as well as a reduction in early adverse events with the administration of higher doses.¹⁴ Such high doses of clopidogrel, when systematically used in pre-treatment, have affected the systematic use—in high-risk ACS—of effective IV-administered antiplatelet agents¹⁵ such as glycoprotein IIb/IIIa (GPIIb/IIIa) platelet receptor inhibitors (eptifibatide, tirofiban, and abciximab), that have been used for a long time in the early treatment of ACS and are today specifically recommended, with a selective approach, for bailout use in the catheterization laboratory or in case of angiographic evidence of intracoronary thrombus.¹⁶

Despite its documented efficacy, clopidogrel has an important limitation, i.e. a high variability of the induced antiplatelet response: ∼25% of treated patients is either ‘resistant’ or under-responsive, showing high levels of residual platelet activity.¹⁷ This clinically relevant phenomenon that exposes patients to an increased risk of death, infarction, and stroke^17,18 can have several causes. High on-clopidogrel platelet reactivity has been mainly observed in patients with platelet hyper-reactivity, such as those with diabetes, elderly patients, or ACS patients undergoing PCI^17,18 in which even the administration of higher loading doses of clopidogrel is not sufficient to ensure clinical efficacy.^19–21 Clopidogrel has a low bioavailability: 85% of the drug is hydrolyzed by plasma esterases into an inactive metabolite, and only 15% is available for biotransformation processes in the liver. The hepatic metabolism that generates the active metabolite is catalyzed by cytochrome P450 isoenzymes, whose genes show a high degree of polymorphism. Therefore, the pharmacological response to clopidogrel is affected by genetic variations:^22,23 some alleles confer a reduced enzyme function, resulting in a reduction in the levels of active drug. The most common polymorphism is CYP2C19*2, an allele variation that involves a functional loss of the enzyme activity and a reduction of up to one-third in the pharmacodynamics and pharmacokinetics of clopidogrel.²⁴ According to the different ability to metabolize clopidogrel, patients can be distinguished into fast, intermediate, and slow metabolizers.²⁵ Even ADP-receptor P2Y₁₂ shows a certain degree of genetic polymorphism that influences the binding site and its metabolic activity, thus modulating clopidogrel biological activity.²⁶

Clopidogrel effect is also influenced by pharmacological interactions that mainly involve cytochrome P450-mediated metabolic processes in the liver. In particular, some proton pump inhibitors (e.g. omeprazole but not pantoprazole) may reduce the efficacy of clopidogrel by inhibiting the activity of isoenzyme CYP2C19;^27–29 likewise, atorvastatin—metabolized by the CYP3A4—may reduce the active concentration of the drug and the associated platelet inhibition.³⁰ In the light of these pharmacokinetic interactions, in 2009 the US Food and Drug Administration (FDA) issued a warning about a reduction in efficacy resulting from concomitant administration of omeprazole and clopidogrel, and recommended to avoid co-administration of clopidogrel and esomeprazole, cimetidine, fluconazole, ketoconazole, felbamate, and fluoxetine.³¹ However, the subsequent Clopidogrel and the Optimization of Gastrointestinal Events Trial study, in which over 3700 patients with coronary heart disease were randomized to receive omeprazole or placebo in addition to their dual antiplatelet therapy, showed a clear reduction in gastrointestinal bleeding events with omeprazole [hazard ratio (HR) 0.34, 95% confidence interval (CI) 0.18–0.64, P < 0.001], with no apparent interaction with cardiovascular events (HR 0.99, 95% CI 0.68–1.44, P = 0.96).³² It must be however acknowledged that the CONGENT study was prematurely terminated due to economic reasons, and therefore the analysis relies only on incomplete data.

The above-described limitations, even more pronounced with the generic formulation of clopidogrel,³³ were the starting point for the investigation of further therapeutic strategies and new, more potent antiplatelet drugs characterized by lower inter-subject variability, thus producing an optimal inhibition of platelet activation^34–36 regardless of the status of gene variations related to hepatic metabolism (Figure 2).^34,37,38

Figure 2

Biotransformation processes of clopidogrel, prasugrel, and ticagrelor. Adapted from Schömig.³⁴

Open in new tabDownload slide

New oral platelet adenosine diphosphate P2Y₁₂ receptor antagonists

In the last few years, research on antiplatelet treatment in ACS has been mainly focused on evaluating the superiority of two new oral inhibitors of platelet receptor P2Y₁₂, prasugrel and ticagrelor, subsequently put on the market, when compared with the standard of care, i.e. clopidogrel.

Prasugrel is a third-generation thienopyridine; it is a selective and irreversible ADP P2Y₁₂ receptor antagonist that, in the light of clinical trial results and by virtue of its pharmacological profile, appears to overcome the limitations associated with clopidogrel treatment.^39,40 The key clinical trial conducted to determine the efficacy and safety of prasugrel when compared with clopidogrel is the Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel-Thrombolysis in Myocardial Infarction (TRITON-TIMI 38) trial: it is a randomized, double-blind, multi-centre study that enrolled 13 608 ACS patients [10 074 with NSTE-ACS and 3534 with ST segment elevation myocardial infarction (STEMI)] with scheduled PCI.⁴¹ Patients were randomized to receive a loading dose of clopidogrel 300 mg or prasugrel 60 mg, followed by a maintenance dose of 75 and 10 mg, respectively, for 6–15 months in association with ASA.⁴¹ It is worth to underline that in this trial the loading dose of study drugs was administered, in some cases, after coronary angiography; therefore, the different pharmacokinetics of clopidogrel and prasugrel should be taken into account. The primary endpoint, which included cardiovascular death, non-fatal myocardial infarction and non-fatal stroke, was observed in 12.1% of patients in the clopidogrel group vs. 9.9% in the prasugrel group (HR prasugrel vs. clopidogrel 0.81; 95% CI 0.73–0.90; P < 0.001); this incidence was significant as of Day 3 and was maintained throughout the follow-up. The observed difference was largely due to a reduction in the incidence rate of non-fatal myocardial infarction (9.7% for clopidogrel vs. 7.4% for prasugrel; P < 0.001), while no differences were reported in the rates of stroke and cardiovascular death.⁴¹ The prasugrel group also showed a significant reduction in urgent revascularizations and stent thrombosis, which was independent of the type of stent used (drug-eluting stent or conventional stent).⁴²

In this study, the primary safety endpoint was represented by major bleeding, which was observed in 2.4% of patients in the prasugrel group vs. 1.8% in the clopidogrel group (HR prasugrel vs. clopidogrel 1.32; 95% CI 1.03–1.68; P = 0.03), with a 32% increase in the RR of major bleeding; moreover, higher rates of life-threatening bleeding (1.4 vs. 0.9%; P = 0.01) and fatal bleeding (0.4 vs. 0.1%; P = 0.002) were observed in the prasugrel group.^41,43 Despite the increase in bleeding risk, the net clinical benefit—that included all-cause mortality, ischaemic events and major bleeding—was in favour of prasugrel. A post hoc analysis identified three subsets of patients in which the net clinical benefit (protection against ischaemia and bleeding risk) of prasugrel was unfavourable: patients with a history of stroke or transient ischaemic attack (TIA) (which represent an absolute contraindication to prasugrel due to the significant increase in major and fatal bleeding events reported in this subset of TRITON-TIMI 38 patients); patients of ≥75 years of age and weighing <60 kg (a relative contraindication).^41,44

Ticagrelor is the first of a new class of direct-acting inhibitors of the ADP-receptor P2Y₁₂.^33,45,46 Unlike thienopyridines (ticlopidine, clopidogrel, and prasugrel) that bind in a covalent manner to and permanently inactivate platelet receptors, ticagrelor reversibly binds to these receptors: it does not prevent ADP-receptor binding, but blocks receptor activation and the resulting conformational change that leads to protein G activation.^33,45,46 Moreover, unlike clopidogrel and prasugrel, ticagrelor is not a pro-drug and does not require biotransformation processes in the liver to produce the active metabolite.^33,45,46

The efficacy and safety of ticagrelor were evaluated in the PLATelet inhibition and patient Outcomes(PLATO) clinical trial,⁴⁷ a randomized, double-blind trial comparing ticagrelor (180 mg loading dose followed by 90 mg twice daily) with clopidogrel (300–600 mg loading dose followed by 75 mg daily), both administered in addition to ASA, in 18 624 ACS (STEMI and NSTE-ACS) patients enrolled within 24 h from the onset of symptoms. Unlike the TRITON-TIMI 38 trial, where patients were scheduled for PCI, in the PLATO trial ACS patients were liable to undergo any therapeutic strategy (either invasive or conservative) (Table 1).

The composite primary endpoint included cardiovascular death, myocardial infarction, and stroke; the primary safety endpoint was the occurrence of major bleeding events [including those associated with coronary artery bypass graft (CABG) procedures].⁴⁷ In both groups, that were followed up for a period of 12 months, the first dose of study drug was administered after a median of 11 h from the onset of symptoms; in the clopidogrel group, 80% of patients received a loading dose of at least 300 and 20% at least 600 mg.⁴⁷ At 12 months, the primary endpoint was observed in 9.8% of ticagrelor-treated patients vs. 11.7% of clopidogrel-treated patients (HR 0.84; 95% CI 0.77–0.92; P < 0.001) (Table 2); the difference was evident after the first 30 days of treatment and was maintained throughout the follow-up. The incidence rates of the individual components of the composite endpoint were as follows (ticagrelor vs. clopidogrel): myocardial infarction 5.8 vs. 6.9% (P = 0.005); cardiovascular death 4.0 vs. 5.1% (P = 0.001); all-cause mortality 4.5 vs. 5.9% (P < 0.001); only the incidence of stroke did not differ significantly in the two groups (1.5 vs. 1.3%; P = 0.22), with a similar number of ischaemic events but more haemorrhagic strokes with ticagrelor, although the difference was not statistically significant (0.2 vs. 0.1%; P = 0.10).⁴⁷ It should be noted that the PLATO trial is one of the few clinical studies conducted in ACS in the last 15 years where a reduction in cardiovascular deaths was reported; this result was observed both in the general population and in the population receiving invasive strategies (Figure 3), and can be attributed to the antiplatelet efficacy of the treatment, as well as to its binding reversibility and to possible adenosine-mediated pleiotropic effects.^48–51

Figure 3

Incidence of cardiovascular mortality in the Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel-Thrombolysis in Myocardial Infarction 38 and PLATelet inhibition and patient outcomes studies and in the PLATelet inhibition and patient outcomes INVASIVE analysis. Blue: prasugrel, yellow: ticagrelor, and red: clopidogrel. P < 0.05 for ticagrelor vs. clopidogrel for both comparisons.

Open in new tabDownload slide

Regarding the safety endpoints, the ticagrelor group reported a higher rate of major bleeding events not related to CABG (2.8 vs. 2.2%, P = 0.03, according to TIMI criteria; 4.5 vs. 3.8% according to the PLATO classification); as regards total fatal bleeding events, similar rates were observed in both groups (ticagrelor and clopidogrel), although there was a higher incidence of fatal intracranial bleeding events in the ticagrelor group (11/9235 vs. 1/9186 patients).^47,52

The PLATO study reported some adverse effects in ticagrelor-treated patients: an increase in the frequency of dyspnoea (13.8 vs. 7.8%) resulting in treatment discontinuation in five patients only, while in most cases it was a self-limiting effect;^47,53 ECG changes, such as ventricular pauses detected during Holter monitoring, which were not accompanied by an increased frequency of syncope or pacemaker placement and were limited to the first treatment week; a significantly higher increase in serum uric acid and creatinine.⁴⁷

To date, 36 sub-analyses have been conducted on PLATO trial data, of which 11 post hoc analyses and 25 pre-specified analyses, while 34 sub-analyses have been conducted for the TRITON-TIMI 38 trial (10 post hoc and 24 pre-specified analyses).

Outline and scope of the paper

This paper has been issued as a result of a meeting held in Florence on March 12 and 13, 2013, by 80 Italian experts in antiplatelet therapy in ACS (see list in Appendix) who were members of the Associazione Nazionale Medici Cardiologi Ospedalieri (ANMCO, Italian Association of Hospital Cardiologists) and the Società Italiana di Cardiologia Invasiva (SICI-GISE, Italian Society of Invasive Cardiology). After selecting 10 subgroups of patients, under the supervision of two coordinators for each subgroup, the evidence available for each antiplatelet drug (intravenous or oral treatment) was largely discussed for each subgroup and a score was assigned (ranging from 0 to 3). The score was given according to the scientific quality of the evidence available for each compound, without expressing a preference or providing indications for the administration of a specific drug or class of drugs; such recommendations are typically stated in the international guidelines,^54,55 which should be used as the reference guidance in the clinical practice and to which the present document represents a propedeutic and ancillary support. The purpose of this paper is to summarize and organize the large amount of data recently published on the antiplatelet therapy currently available and used in the hospital management of ACS; for each compound or class of drugs and for each subgroup, strong evidence and/or areas of uncertainty that warrant further research have been highlighted. When a higher score has been given to a drug compared with another of the same class, this does not mean that the drug is superior to another, but that the study or sub-study generating the evidence on that compound has shown better criteria and features from a scientific point of view. Therefore, while the ‘opinion-based’ nature of this consensus—which has a focus on Italian clinical practice—should be acknowledged, this opinion is based on the scientific robustness of the published and presented data, and not on a more or less established clinical experience, especially when the evaluation concerns recently approved drugs.

It should also be noted that, regarding the two new oral inhibitors of platelet receptor P2Y₁₂ (prasugrel and ticagrelor), no head-to-head studies have been conducted to date (with the exception of small pharmacodynamics studies) and any direct comparison would be inappropriate from a biostatistical point of view, also because most evidence comes from clinical trials that are completely different from one another in terms of study design, inclusion criteria and therapeutic strategies.⁵⁶ The ongoing Intracoronary Stenting and Antithrombotic Regimen: Rapid Early Action for Coronary Treatment -5 trial will perform a head-to-head comparison of the two drugs in terms of death, myocardial infarction and stroke at 12 months in 4000 patients with ACS and planned invasive strategy.⁵⁷

Given the increasingly high availability of antiplatelet drugs, the cost of new drugs and the genericization of old compounds, it is highly desirable that expert groups identify subgroups of patients for which the use of innovative drugs is recommended in consideration of the observed clinical benefit. In this context, the cost-effectiveness of the drugs used should be always assessed, and therefore it is important to identify patients who take most advantage of an innovative intervention. Although all subgroup analyses generate evidence and hypotheses that can be used to identify an individualized therapy based on the clinical and/or procedure-related characteristics of patients, such an attractive therapeutic strategy is still under discussion and can be confirmed only by prospective studies conducted in the selected subgroup with the purpose of validating a given hypothesis.

Of course, a different approach to this clinical issue may have been also used, i.e. to recommend a single agent, and leave other available drugs for very specific and narrow subgroups. While such approach is not without its merits, the Members of the Panel preferred, after some deliberation, to focus on the use of different agents in given subgroups of patients.

Non-ST elevation acute coronary syndrome

In the last decades, the proportion of NSTE-ACS patients undergoing coronary angiography, either as an initial invasive strategy or following exacerbation in patients initially managed by conservative treatment, has increased considerably.^58–60 In Italy, the use of coronary angiography in ACS patients appears to be related more to the presence of a cath lab in the centre where a patient has been institutionalized than to his/her estimated ischaemic risk profile.⁶¹ In 2010, according to the data of the Italian registry ‘ANMCO BLITZ-4 Quality’, 63% of patients institutionalized with an ACS received coronary angiography in hospitals that were not equipped with a cath lab, when compared with 89% of ACS patients in hospitals with a round-the-clock cath lab.⁶² In addition to the different strategy, it should be considered that, according to whether a cath lab is available or not, there is also a marked variation in the time interval between diagnosis and invasive treatment.

Although pre-treatment with antiplatelet agents is indicated by the current guidelines with a Class I recommendation across the ACS spectrum,^54,55 the role of early administration of a loading dose of thienopyridines—in terms of risk-benefit ratio—is still debated, especially in NSTE-ACS (although the panel has confirmed the appropriateness of an early administration of ASA regardless of the strategy initially chosen, with a score of 3). In this respect, an advantage of the early administration of clopidogrel was shown by the PCI-CURE study, where patients of the CURE study undergoing PCI were also treated with clopidogrel plus ASA or ASA alone in the first 30 days after PCI. As previously mentioned, the study objective was to determine whether pre-treatment with clopidogrel, followed by a long-term therapy, reduced the composite primary endpoint of cardiovascular death, myocardial infarction, and urgent revascularization.¹³ The composite primary outcome measure, within 30 days following PCI, was lower with clopidogrel plus ASA (4.5 vs. 6.4%). In terms of risk, the trial showed an RR reduction of 30% up to 30 days after PCI with clopidogrel plus ASA in the composite primary outcome measure, and a reduction of 25% in the risk of cardiovascular death and infarction in the 12 months following PCI. The addition of clopidogrel to ASA did not increase the risk of severe bleeding (1.6 vs. 1.4%).¹³ Moreover, a pre-specified analysis of the clopidogrel for the reduction of events during observation (CREDO) trial showed a clinical benefit with this therapy only when the time interval between diagnosis and coronary angiography exceeded 6 h.⁶³ A meta-analysis of the PCI-CURE, CREDO, and PCI-CLARITY (PCI-Clopidogrel as Adjunctive Reperfusion Therapy) trials showed an advantage for the upstream vs. incath lab treatment in terms of pre-PCI infarction or post-PCI mortality and infarction.⁶⁴ Recently, another meta-analysis of six randomized clinical trials and nine observational studies that included a total of ∼38 000 patients with ACS or stable chronic angina reported a benefit associated with clopidogrel pre-treatment in terms of major cardiac events [9.8 vs. 12.3%; odds ratio (OR) 0.77; 95% CI 0.66–0.89; P < 0.001], although there was no effect on overall mortality (a benefit that was never reported in the studies included in the meta-analysis).⁶⁵In the light of these data, considering the current guidelines and the large clinical experience, the panel assigned a score of 2 to upstream clopidogrel (pre-treatment) and a score of 0 to downstream treatment (incath lab administration) in NSTE-ACS.

Patients with a diagnosis of NSTE-ACS are the majority of the population enrolled in the key trials on new oral inhibitors of platelet receptor P2Y₁₂ (PLATO and TRITON-TIMI 38).

In a retrospective analysis of the PLATO trial, among the 11 080 patients categorized as NSTE-ACS at randomization, 5366 (48.4%) patients were managed without revascularization within the first 10 days. Regardless of revascularization or not, ticagrelor consistently reduced the primary outcome (HR 0.86 vs. 0.85, interaction P = 0.93), and all-cause death (HR 0.75 vs. 0.73, interaction P = 0.89) with no significant increase in overall major bleeding.⁶⁶

The TRITON-TIMI 38 trial, too, did not reveal a heterogeneity between the STEMI and the NSTE-ACS groups of patients,⁶⁷ and therefore the benefit associated with prasugrel when compared with clopidogrel in the reduction of the primary endpoint (mainly attributed to a dramatic reduction in non-fatal myocardial infarction) could be extrapolated to the NSTE-ACS subgroup. More specifically, the primary endpoint was significantly reduced by prasugrel in the overall NSTE-ACS population (HR 0.82, 95% CI 0.73–0.93, P = 0.002) as well as in UA and in NSTEMI patient subgroups (interaction P-value = 0.39). Although non-CABG TIMI major bleeding was increased with prasugrel when compared with clopidogrel (HR 1.40, 95% CI 1.05–1.88, P = 0.02), there was a net clinical benefit in patients assigned to prasugrel (HR 0.89, 95% CI 0.80–1.00, P = 0.043), which was consistent for UA and NSTEMI subgroups (interaction P-value = 0.84 and 0.72).⁶⁷

Regarding pre-treatment, the benefits associated with early ticagrelor use have been demonstrated by the PLATO trial, where ACS patients could be treated even before their coronary anatomy had been assessed and irrespective of the selected type of strategy.⁴⁷ This was not possible in the TRITON-TIMI 38 trial due to the study design, since randomization was performed in the cath lab in clopidogrel-naive patients, when the diagnosis was NSTE-ACS.⁴¹In the light of such data and because of the different designs of the key trials (PLATO and TRITON-TIMI 38), the panel assigned a score of 3 to ticagrelor (administered as a pre-treatment or in the cath lab) and a score of 3 to prasugrel only when administered after coronary angiography in naïve patients for whom a PCI is planned, with the known limitations of use, and assigned a score of 0 to upstream prasugrel. Assignment of this score is also based on the evaluation of the ACCOAST (A Comparison of Prasugrel at PCI or Time of Diagnosis of Non-ST Elevation Myocardial Infarction) trial, where prasugrel administered at a loading dose of 30 mg before coronary angiography, followed by another dose of 30 mg in case of PCI, was compared with a standard loading dose of prasugrel (60 mg) administered in the cath lab after assessment of coronary anatomy.⁶⁸ Indeed in the ACCOAST trial, the rate of the primary efficacy endpoint, a composite of death from cardiovascular causes, myocardial infarction, stroke, urgent revascularization, or GPIIb/IIIa inhibitor rescue therapy (GPIIb/IIIa bailout) through Day 7, did not differ significantly between the two groups (HR with pre-treatment, 1.02; 95% CI 0.84–1.25; P = 0.81). The rate of the key safety endpoint of all TIMI major bleeding episodes, whether related or not related to CABG, through Day 7 was increased with pre-treatment (HR 1.90; 95% CI 1.19–3.02; P = 0.006).⁶⁸

As regards GPIIb/IIIa inhibitors, there is a strong evidence in terms of reduction of ischaemic events when used in addition to ASA and clopidogrel, and they can be considered a valuable pharmacological treatment that can be ideally used in patients with a low bleeding risk and a high risk of ischaemic events during PCI.⁶⁹ To date, however, we have no valid clinical data available on the addition of GPIIb/IIIa inhibitors to ASA and new oral inhibitors of platelet receptor P2Y₁₂, although there are mechanistic grounds to assume that an intravenously administered antiplatelet agent could be beneficial in NSTE-ACS patients, since even the new oral antiplatelet drugs show a remarkable delay in achieving optimal platelet inhibition.⁷⁰ The systematic upstream use of GPIIb/IIIa inhibitors was studied as add-on to ASA and clopidogrel (the latter administered with a varying time schedule left to the investigator's discretion) in the EARLY-ACS (Early Glycoprotein IIb/IIIa Inhibition in Non-ST-Segment Elevation Acute Coronary Syndrome) trial; in this study, ∼9500 NSTE-ACS patients, in whom an invasive approach was planned, received an early or delayed treatment with eptifibatide.⁷¹ The primary endpoint (death, myocardial infarction, urgent revascularization or thrombotic complications during PCI at 96 h) was comparable with early vs. delayed eptifibatide (9.3 vs. 10.0%; OR 0.92; 95% CI 0.80–1.06; P = 0.23), with no significant interactions in the analysed subgroups, including diabetic patients or patients with positive troponin. On the contrary, the frequency of major bleeding events was higher in patients who had received early eptifibatide (2.6 vs. 1.8%; OR 1.42; 95% CI 1.97–1.89; P = 0.015).⁷¹ In the Acute Catheterization and Urgent Intervention Triage Strategy Trial trial, GPIIb/IIIa delayed (i.e. only during PCI) vs. acute administrations were compared in 9207 ACS patients who had been pre-treated with thienopyridines in 64% of cases.⁷² Delayed treatment was associated with a lower incidence of non-CABG-related major bleeding events at 30 days when compared with upstream administration (4.9 vs. 6.1%; RR 0.80; 95% CI 0.67–0.95; P = 0.009), with no differences in the rate of adverse ischaemic events (7.9 vs. 7.1%; RR 1.12; 95% CI 0.97–1.29; P = 0.13).⁷³ To date, the upstream use of GPIIb/IIIa inhibitors in patients at high risk of ischaemic events (e.g. with recurrent angina) and low bleeding risk is regarded as a reasonable strategy, especially in clopidogrel-naive patients (who are highly likely to have surgical anatomy) who cannot undergo coronary angiography and possibly a PCI within a short time, while incath lab use is reserved to patients with high-risk procedural characteristics (e.g. intracoronary thrombus). In the light of such data, the panel assigned a score of 1 to the class of GPIIb/IIIa inhibitors for both upstream and downstream use.

ST elevation acute myocardial infarction

Intracoronary thrombus formation in STEMI patients is a dynamic process.⁷³ Platelet and fibrin contents are closely dependent on the ischaemic time, resulting in a positive correlation with fibrin content and a negative correlation with platelet content.⁷³ This may directly affect the efficacy of the antiplatelet drugs used in STEMI. Such agents, in association with coronary revascularization in STEMI patients, have the purpose to facilitate myocardial reperfusion before starting the reperfusion treatment, or to obtain an early appropriate therapeutic level of active drug at the time of mechanical reperfusion. In this setting, it should also be noted that agents such as oral antiplatelet drugs—by virtue of their route of administration and relatively low bioavailability—are given with the only purpose of reducing ischaemic relapses when an STEMI diagnosis is established.

Although there are no randomized clinical trials specifically conducted on the use of clopidogrel in patients with a diagnosis of STEMI, a meta-analysis of clinical trials⁷⁴ has shown that the administration of 300 mg clopidogrel approximately a median time of 2 h before primary PCI was associated with a lower incidence of mortality and re-infarction. Moreover, retrospective data from large registries or post hoc analyses of randomized clinical trials have documented that pre-treatment with clopidogrel 300 mg in STEMI patients produces a beneficial effect in terms of composite ischaemic endpoint and mortality.^75,76 Likewise, a post hoc analysis of the randomized clinical trial Harmonizing Outcomes With Revascularization and Stents in AMI (HORIZONS-AMI), suggested that such benefit may increase with a loading dose of 600 mg without producing any further increase in major bleeding events.⁷⁷ Therefore, the panel assigned a score of 1 to clopidogrel in STEMI because of the paucity of specific published data, and due to the fact that the available data show a superiority of the new inhibitors of platelet receptor P2Y₁₂ over clopidogrel in this subgroup of patients, when both are used in association with ASA (the panel assigned a score of 3 to the latter association).

When clopidogrel is administered due to contraindications to or unavailability of the new P2Y₁₂ inhibitors, it is suggested to use a loading dose of 600 mg in patients with an STEMI diagnosis for whom a primary PCI is planned; if a thrombolytic drug is used, it is recommended to administer clopidogrel at a loading dose of 300 mg (there are no data available on the early association of a thrombolytic drug and new oral inhibitors of platelet receptor P2Y₁₂).

The new inhibitors of platelet receptor P2Y₁₂ should be used as first-choice drugs in the STEMI setting, as suggested by the European guidelines,⁵⁵ as they produce better long-term clinical outcomes when compared with clopidogrel.⁷⁸ It should be noted that the STEMI population is a pre-specified subgroup of the PLATO trial and a cohort of the TRITON-TIMI 38 trial (where the calculation of sample size was based on NSTE-ACS patients, and the number of STEMI patients was added subsequently with a calculation based on the incidence of STEMI in the real world). The TRITON-TIMI 38 trial admitted randomization, with coronary anatomy unknown, to prasugrel 60 mg vs. clopidogrel 300 mg in patients with a diagnosis of STEMI and onset of symptoms <12 h who were scheduled for primary PCI.⁷⁹ Overall, the mean time between the onset of symptoms and drug administration in the STEMI cohort of the TRITON-TIMI 38 trial (∼3500 patients) was 228 min, and the mean time between the onset of symptoms and primary PCI was 252 min; only 31% of these patients could benefit from an upstream treatment with prasugrel,⁷⁹ with an advantage over clopidogrel in the composite primary endpoint of cardiovascular death, non-fatal infarction and non-fatal stroke (10 vs. 12.4%; P = 0.02) and stent thrombosis (1.6 vs. 2.8%; P = 0.02) and without a relative increase in major bleeding events.⁷⁹ It should be noted that the benefits observed in the STEMI cohort of the TRITON-TIMI 38 trial were driven by subjects undergoing secondary PCI (over 12 h after the onset of symptoms), ∼30% of the cohort, in whom randomization was performed after coronary angiography as in the general population (HR 0.65; P = 0.01); on the other hand, no benefits were observed with prasugrel in the primary endpoint in STEMI patients undergoing primary PCI (HR 0.87; P = 0.26). Although this sub-analysis may generate clinical considerations, it should be interpreted bearing in mind that there is no interaction between the type of STEMI population enrolled in the study and treatment effect.⁸⁰ In addition when periprocedural MI was excluded, the efficacy of prasugrel remained consistent among primary and secondary PCI patients at 30 days (HR 0.53; 95% CI 0.34–0.81 vs. HR 0.44; 95% CI 0.22–0.88; P-interaction = 0.68) and 15 months (HR 0.76; 95% CI 0.56–1.03 vs. HR 0.75; 95% CI 0.46–1.21; P-interaction = 0.96).⁸⁰

In the PLATO trial, over 7000 patients with an STEMI diagnosis were randomized to ticagrelor 180 mg or clopidogrel 300–600 mg; 70% of patients were undergoing primary PCI (within 12 h from randomization) and the mean time between the onset of symptoms and randomization was ∼5 h.⁸¹ The reduction observed in the primary endpoint in the ticagrelor group vs. the clopidogrel group (10.8 vs. 9.4%; HR 0.87; 95% CI 0.75–1.01; P = 0.07) was consistent with that observed in the total population, with no interactions between STEMI diagnosis and the treatment arm (P for the interaction = 0.29). In addition, ticagrelor was associated with a reduction in some of the secondary endpoints including myocardial infarction (HR 0.80; P = 0.03), all-cause mortality (HR 0.82; P = 0.05), and stent thrombosis (HR 0.66; P = 0.03); however, an increased incidence of stroke was seen in the ticagrelor arm (1.7 vs. 1.0%; HR 1.63; 95% CI 1.07–2.48; P = 0.02), although major bleeding events were not increased (HR 0.98; P = 0.76).⁸¹ In this subgroup, too, there were patients with late reperfusion (secondary PCI), and according to some FDA analyses, the benefit observed with ticagrelor in terms of all-cause mortality was particularly evident in patients with STEMI and with the onset of symptoms exceeding 10 h (an arbitrary and not pre-specified cut-off).⁸² In this respect, the benefits observed with ticagrelor in STEMI patients do not appear to be related to the rapid onset of action of the drug in the acute phase of reperfusion, but to the prevention of recurrent vascular events starting from the post-acute phase, as suggested by the angiographic study⁸³ and the ECG study of the STEMI subgroup in the PLATO trial.⁸⁴ However, it should be noted that the PLATO sub-study on myocardial reperfusion has important methodological limitations: although the decision to use the degree of resolution of ST elevation is widely justified by the literature, the median time interval between baseline ECG and the ECG performed to assess the degree of ST resolution was 3 days, while other studies conducted to validate the prognostic value of this marker were systematically using immediate post-PCI ECG.⁸⁴ The randomized ATLANTIC (Administration of Ticagrelor in the cath Lab or in the Ambulance for New ST elevation myocardial Infarction to open the Coronary artery) trial had the objective to determine whether ticagrelor could be a valuable aid in the pre-hospital phase of an STEMI to be treated with primary PCI.⁸⁵ The coprimary endpoints were the proportion of patients who did not have a 70% or greater resolution of ST-segment elevation before PCI and the proportion of patients who did not have TIMI flow grade 3 in the infarct-related artery at initial angiography. Secondary endpoints included the rates of major adverse cardiovascular events and definite stent thrombosis at 30 days.⁸⁵ The two coprimary endpoints did not differ significantly between the pre-hospital and in-hospital groups. The rates of definite stent thrombosis were lower in the pre-hospital group than in the in-hospital group (0 vs. 0.8% in the first 24 h; 0.2 vs. 1.2% at 30 days). Rates of major bleeding events were low and virtually identical in the two groups, regardless of the bleeding definition used.⁸⁵ Although the two studies on new inhibitors of platelet receptor P2Y₁₂ (TRITON-TIMI 38 and PLATO) cannot be directly compared also due to the heterogeneity of the STEMI population enrolled⁸⁶ (Table 3), it should be noted that there seems to be a difference in the timing of effect of the two drugs (prasugrel and ticagrelor) in the STEMI subgroup, as shown by Kaplan–Meier curves: in the PLATO trial, the curves diverge not in the acute phase but starting from the first month after randomization, while in the TRITON-TIMI 38 trial the majority of the effect is observed in the acute phase and the curves do not diverge further after that phase. This may be related to the different study designs, inclusion criteria, associated antiplatelet therapy, randomization timing, or pharmacokinetic profiles of the two drugs, as suggested by recent comparative mechanistic studies of prasugrel vs. ticagrelor in addition to bivalirudin in patients with a diagnosis of STEMI.⁸⁷In consideration of the data presented and after ample discussion, the majority of the panel assigned a score of 3 to prasugrel (except when used in patients with a history of stroke/TIA and/or aged ≥75) and a score of 2 to ticagrelor for the STEMI subgroup; however, this statement is opinion based. The rationale for the use of GPIIb/IIIa inhibitors in STEMI is very strong, if we consider the high platelet aggregation, the prominent platelet component in the intracoronary thrombus during the first hours after the onset of symptoms,⁷³ and the reduced gastric absorption of orally administered drugs that is associated with hemodynamic instability and adrenalin production with subsequent gastric vasoconstriction.⁸⁸ Abciximab is certainly the GPIIb/IIIa inhibitor that has been more widely tested in STEMI, with over 20 published randomized clinical trials when compared with ∼10 trials for tirofiban and 4 for eptifibatide. In the context of primary PCI, several meta-analyses have found that the use of abciximab is associated with a reduction in 1-year mortality^89,90 or in death and myocardial infarction after 3 years of follow-up.⁹¹ Such benefits also appear to be associated with the use of other GPIIb/IIIa inhibitors, which show a substantial equivalence to abciximab in terms of death, re-infarction, TIMI flow and ST resolution.⁹² The administration of GPIIb/IIIa inhibitors in the pre-hospital phase in STEMI, especially in addition to high doses of clopidogrel, is still a controversial issue; while the Abciximab before Direct Angioplasty and Stenting in Myocardial Infarction Regarding Acute and Long-term Follow-up trial⁹³ supported pre-hospital administration of abciximab, the results of the large randomized FINESSE trial (Facilitated Intervention with Enhanced Reperfusion Speed to Stop Events)⁹⁴ did not support the routine, early administration of GPIIb/IIIa inhibitors. On the contrary, an angiographic analysis of the EUROTRANSFER registry documented an improvement in the incidence of TIMI 2 and 3 flow in the group of STEMI patients treated with abciximab.⁹⁵ These angiographic results also translated into an improvement of long-term survival, especially in STEMI patients with early presentation and not high risk.⁹⁶ Another GPIIb/IIIa inhibitor, tirofiban, administered in the pre-hospital phase in association with 600 mg clopidogrel, ASA and heparin, has shown to produce benefits when compared with placebo in mean residual ST deviation 1 h after primary PCI in the randomized Ongoing Tirofiban In Myocardial infarction Evaluation-2 trial⁹⁷; this benefit was even greater in patients with a short-time period between the onset of symptoms and randomization.⁹⁸ Therefore, the panel assigned a score of 3 to abciximab, a score of 2 to tirofiban and a score of 1 to eptifibatide in STEMI patients.

Conservative treatment

Recent international registries have shown that ∼40% of patients hospitalized with a diagnosis of ACS receive a conservative treatment^99–102 (Figure 4). Among these patients, two distinct populations can be identified: patients who do not even receive coronary angiography due to their high clinical risk (frail patients with several co-morbidities) or very low risk; patients who are not revascularized after coronary angiography due to their extremely low anatomic risk (subcritical coronary artery disease or obstructive disease of a secondary vessel) or, more rarely, due to the presence of a severe and extensive obstructive coronary artery disease not eligible for revascularization. While the latter population has a low mortality rate (∼2–2.5%) and remains stable over time, the first population appears to show a progressive numerical reduction and a simultaneous increase in mortality over the years (up to 7–9% in the in-hospital phase).^99–102 In addition to showing a higher risk profile and no revascularization, this group of patients appears to receive sub-optimal pharmacological treatment, with an under-use of guideline-recommended antithrombotic drugs at the time of hospital discharge,^99–102 which may further affect the high incidence of long-term fatal events. Despite the lack of ad hoc randomized trials on the use of ASA in this subgroup of patients, in consideration of secondary prevention studies, of the biological plausibility and of the large clinical experience accumulated over the years, the panel assigned a score of 3 to the use of ASA in all conservative strategy settings.

Figure 4

Percentage of patients undergoing conservative treatment in recent randomized clinical trials, observational studies, or registries. RCT, randomized controlled trial.

Open in new tabDownload slide

As previously described, treatment with clopidogrel in ACS was validated by the CURE trial, which was primarily designed to evaluate a ‘conservative’ approach; in fact, study centres preferentially included cardiology units that did not tend to perform early angiography and coronary revascularization.⁹ In the CURE trial, however, the greatest benefit obtained with clopidogrel when compared with placebo was seen in the population undergoing coronary revascularization, while among patients medically managed the use of clopidogrel was associated with a 20% RR reduction of the primary endpoint.¹⁰³ These findings have been recently confirmed in a large, community-based cohort of patients who were medically managed after NSTE-ACS, where clopidogrel use was associated with a lower risk of death and MI, particularly among patients with NSTEMI.¹⁰⁴In consideration of the above data and of the large clinical experience accumulated over the years, and after ample discussion about the attention to be reserved to patients at high risk of bleeding who receive conservative treatment, the panel assigned clopidogrel a score of 2 when the coronary anatomy has not been evaluated, and a score of 1 in case of conservative therapy when the coronary anatomy is known.

As regards the use of new oral inhibitors of platelet receptor P2Y₁₂ in the conservative strategy, data on ticagrelor are available from a pre-specified analysis of the PLATO trial¹⁰⁵ that analysed patients (28% of the total study population) initially scheduled for conservative treatment (although ∼25% of them subsequently received percutaneous or surgical revascularization). In this analysis, the incidence of the primary endpoint (death, infarction, and stroke) was lower in the ticagrelor group when compared with the clopidogrel group (12.0 vs. 14.3%; HR 0.85; 95% CI 0.73–1.00; P = 0.04), with a significant reduction also in all-cause mortality (6.1 vs. 8.2%; HR 0.75; 95%CI 0.61–0.93; P = 0.01).¹⁰⁵In the light of such evidence, the panel assigned a score of 3 to ticagrelor in all conservative strategy settings.

As regards prasugrel, efficacy data in the conservative treatment setting are provided by the recent Targeted Platelet Inhibition to Clarify the Optimal Strategy to Medically Manage Acute Coronary Syndromes (TRILOGY ACS) trial.¹⁰⁶ In this study, prasugrel was not associated with any clinical benefits when compared with clopidogrel: at 30 months, 13.9% of prasugrel-treated patients and 16% of clopidogrel-treated patients experienced any of the events pre-specified in the primary endpoint (cardiovascular death, myocardial infarction, and stroke) (HR 0.91; 95% CI 0.79–1.05; P = 0.21). Of note, a post hoc analysis conducted to evaluate the effect of the drug in the first 12 months when compared with the subsequent period (up to 30 months of follow-up) showed a superiority trend for prasugrel: 0.99 (0.84–1.16) vs. 0.72 (0.54–0.97) (P = 0.07).¹⁰⁶ The reason for this potential delayed effect, described in other analyses of ACS patients undergoing conservative treatment,^107,108 is not known and could be investigated in other analyses or trials with longer follow-up periods. As regards safety, the major bleeding events reported in patients under the age of 75 and defined using TIMI criteria did not differ between treatment groups: 2.1% with prasugrel vs. 1.5% with clopidogrel (HR 1.31; 95% CI 0.81–2.11; P = 0.27).¹⁰⁶ Such safety data were also confirmed in the TRILOGY ACS population over the age of 75 (HR 1.03; 95%CI 0.86–1.22).¹⁰⁹ A recent publication of the pre-specified analysis of TRILOGY ACS patients undergoing coronary angiography (43% of the population under 75 years of age) showed a reduction in the primary endpoint favouring prasugrel over clopidogrel [122/1524 (10.7%) vs. 159/1561 (14.9%); HR 0.77; 95% CI 0.61–0.98; P = 0.032),¹¹⁰ thus confirming prasugrel-induced benefit in ACS patients with known coronary anatomy.¹¹¹In consideration of the paucity of superiority data of prasugrel when compared with clopidogrel in patients receiving conservative treatment, the panel assigned a cautionary score of 1 in all possible scenarios (known or unknown coronary anatomy).

Coronary angioplasty

Percutaneous coronary intervention is the most commonly used revascularization procedure in ACS and has dramatically improved prognosis in ACS patients.^112,113 Considering that ACS patients are characterized by high thrombin production and platelet aggregation, and since PCI with intracoronary stent placement is a procedure at high thrombotic risk, it can be realized that a maximal antithrombotic effect should be achieved during invasive procedures. Antiplatelet therapy and PCI are two rapidly evolving fields of research and have been evaluated in a number of randomized clinical trials during the last few years, with continuous updates in the international guidelines on ACS. In this context, several issues are still open such as drug combinations, timing of administrations, dosage and duration of pharmacological treatment.¹¹⁴

A meta-analysis of the Antithrombotic Trialists' Collaboration, that included 287 studies in high-risk patients with cardiovascular disease, has shown that ASA reduces the incidence of death, myocardial infarction, and stroke by 25%.¹¹⁵ Such reduction in the rate of in-hospital mortality and stroke with pre-procedural ASA was also confirmed in a recent patient series undergoing PCI.¹¹⁶ The ongoing GLOBAL LEADERS multi-centre trial has been designed to evaluate whether chronic treatment with ASA can be reduced after stent placement by adding a more effective and prolonged antiplatelet treatment with a new inhibitor of platelet receptor P2Y₁₂.¹¹⁷Based on the current knowledge, the panel assigned a score of 3 to the use of ASA in ACS patients undergoing PCI.

The Clopidogrel Optimal Loading Dose Usage to Reduce Recurrent Events/Optimal Antiplatelet Strategy for Interventions-7 trial was specifically designed to evaluate the optimal doses of clopidogrel and ASA in ACS patients undergoing PCI within 72 h from hospital admission.¹¹⁸ Recent data suggested that a double dose of clopidogrel (600 mg loading dose followed by 150 mg/day for 1 week) was able to produce a more rapid and potent effect when compared with the standard dose (300 mg loading dose followed by 75 mg/day), potentially resulting in a higher net clinical benefit. The study, conducted in 597 coronary care units in 39 countries, enrolled >25 000 patients. Among 17 000 patients undergoing PCI (one of the 13 pre-specified analyses of the CURRENT trial), a 15% reduction was observed with double-dose clopidogrel in the primary endpoint of cardiovascular death, myocardial infarction, and stroke (HR 0.86; 95% CI 0.74–0.99; P = 0.039),¹¹⁹ although significance was higher than that required and pre-specified to demonstrate superiority (P = 0.01). This benefit was mainly driven by a 22% reduction in myocardial infarction. Moreover, a 42% reduction was observed in the risk of stent thrombosis. Double-dose clopidogrel did not produce any increase in fatal or intracranial haemorrhages, although a significant increase was reported in major bleeding events.¹¹⁸ The study also demonstrated that there was no difference between low doses (75–100 mg/day) and high doses (300–325 mg/day) of ASA in terms of efficacy and safety.¹¹⁸For these reasons, and considering the documented superiority of new oral inhibitors of platelet receptor P2Y₁₂, the panel recommended 600 mg clopidogrel in ACS patients undergoing PCI, and assigned a score of 1 within this setting.

Following a sub-analysis of the PLATO trial, data were published on the population managed with an invasive strategy (PLATO INVASIVE), that included not only patients undergoing PCI but also those who had initially been scheduled for coronary angiography and/or CABG.¹²⁰ At the time of randomization, an invasive strategy was planned in 72% of patients with ACS enrolled in the trial. Consistently with the general population enrolled in the PLATO trial, the primary endpoint was reported in 19% of the ticagrelor group vs. 10.7% of the clopidogrel group (HR 0.84; 95%CI 0.75–0.94; P = 0.0025), with no difference in terms of major or severe bleeding events.¹²⁰ It should be noted that, in the PLATO trial, clopidogrel and ticagrelor reloading in patients undergoing PCI was admitted and recommended by the study protocol; this suggested a highly heterogeneous population, of which 50% had received clopidogrel in the pre-treatment phase and 20% had received >600 mg clopidogrel in both randomization arms (Table 4). Ticagrelor produced a significant reduction in the incidence of stent thrombosis according to all Academic Research Consortium (ARC) definitions: definite (HR 0.67; 95% CI 0.50–0.90; P = 0.0091); definite or probable (HR 0.75; 95% CI 0.59–0.95; P = 0.017); and definite, probable, or possible (HR 0.77; 95% CI 0.62–0.95).¹²¹ Due to an effect that is likely to be attributed to chance, stent thrombosis was, in absolute terms, significantly reduced in the ticagrelor arm with respect to conventional stents but not to drug-eluting stents (both for definite thrombosis, P = 0.23, and for definite or probable thrombosis, P = 0.65); this difference was not significant as regards P-value for the interaction (P = 0.74), which however, in the same analysis of stent thrombosis, was close to significance (P-value for the interaction = 0.07) as regards the comparison between patients who had received more vs. <600 mg clopidogrel before randomization.¹²¹For these reasons, the panel assigned a score of 2 to the quality of data on ticagrelor use in ACS patients undergoing PCI.

Efficacy data on prasugrel in patients undergoing PCI are provided by the TRITON-TIMI 38⁴¹ trial, which included only this type of population (see previous sections). In the TRITON-TIMI 38 trial, a remarkable reduction in the incidence of stent thrombosis was reported in patients randomized to prasugrel: possible or probable thrombosis, according to ARC definition, was reduced by 52% (1.1 vs. 2.4%; P < 0.001) while definite thrombosis (as established during angiography or post-mortem examination) was reduced by 58% (0.9 vs. 2.0%; P < 0.001). Such data were consistent among patients who had received conventional or drug-eluting stents.^41,42In the light of this specific evidence and after ample discussion, the panel assigned a score of 3 to prasugrel in clopidogrel-naïve ACS patients who do not show the known limitations for use and who are scheduled for a PCI; however, this statement is opinion based.

For a long time, GPIIb/IIIa inhibitors have been used in ACS patients undergoing PCI, and considerable evidence has been produced prior to the introduction of new antiplatelet drugs. Observational studies, randomized trials, and meta-analyses on tens of thousands of patients suggested a benefit with GPIIb/IIIa inhibitors in ACS patients who are to undergo PCI, especially those with positive biomarkers, in terms of mortality, re-infarction and surrogate endpoints.^95,119 When compared with new antithrombotic drugs used in association with PCI, such as bivalirudin, the net clinical benefits of GPIIb/IIIa inhibitors is lower due to an increase in the associated risk of major bleeding events;^72,122,123 therefore, their use in the cath lab has been limited to bailout or to high-risk cases such as an evidence of intracoronary thrombosis.

Recently, news studies raised the question whether bivalirudin does indeed reduce major bleeding in the setting of contemporary primary PCI, compared with UFH, without routine GPIIb/IIIa inhibition and when newer antiplatelet drugs are used. For instance, the large, single-centre, randomized HEAT-PPCI (How Effective Are Antithrombotic Therapies in Primary Percutaneous Coronary Intervention) trial showed no differences in bleeding but an increase in stent thrombosis with bivalirudin compared with heparin, when GPIIb/IIIa inhibitors were used just for bailout in both arms¹²⁴. Accordingly, the BRAVE-4 (Efficacy Study of Combined Prasugrel and Bivalirudin vs. Clopidogrel and Heparin in Myocardial Infarction)¹²⁵ aimed to compare a 60-mg loading dose of prasugrel plus an IV bolus plus infusion of bivalirudin vs. 600 mg clopidogrel plus 70- to 100-IU/kg bolus UFH did not show any difference in the rate of primary composite and secondary endpoints at 30 days.

The ongoing HORIZONS-AMI II trials will further evaluate the association of bivalirudin and new oral P2Y₁₂ inhibitors in STEMI patients undergoing PCI.

Recent studies and meta-analyses have tested the efficacy and safety of different methods of administration of GPIIb/IIIa inhibitors during primary PCI, i.e. peripheral vs. intracoronary administration, with promising results after 30 days in favour of intracoronary administration.^126–132 Afterwards, these two strategies have been evaluated in a large randomized clinical trial, the AIDA STEMI (Abciximab i.v. vs. i.c. in STEMI) trial, in which over 2000 patients were randomized to receive peripheral infusion (bolus plus infusion) vs. intracoronary bolus plus peripheral infusion of abciximab, without reporting any significant difference in terms of survival, re-infarction or heart failure at 90 days (7.0 vs. 7.6%; OR 0.91; 95% CI 0.64–1.28; P = 0.58)¹³³ and at 1 year.¹³⁴ On the contrary, there are favourable data for the locoregional administration of abciximab using a microporous balloon. In the Intracoronary Abciximab and Aspiration Thrombectomy in Patients With Large Anterior Myocardial Infarction (INFUSE-AMI) trial¹³⁵, 450 patients with anterior myocardial infarction appropriately pre-treated with oral antiplatelet agents and bivalirudin were randomized, with a 2 × 2 design, to abciximab by locoregional administration vs. no abciximab and manual trombectomy vs. standard treatment; the trial showed a reduction in infarct size at 30 days in favour of local intracoronary administration (P = 0.03). It should also be noted that intracoronary bolus was not followed by abciximab infusion in the INFUSE-AMI trial, which may explain the low incidence of major bleeding events in the trial. Although administration of the bolus alone (also by peripheral route) is common practice and is based on pharmacokinetic efficacy assumptions and a plausible better safety profile, the panel believes that additional clinical studies are warranted to provide a clear recommendation on this practice. Moreover, although the intracoronary administration strategy needs additional scientific data (probably using a different dose from the peripheral administration or, as previously mentioned, avoiding the infusion in order to minimize bleeding complications) before it is routinely adopted in the clinical practice, it should be emphasized that abciximab is the drug used in most of the studies conducted with this innovative method of administration. Having analysed the amount and quality of data on peripheral administration, the panel assigned a score of 2 to abciximab and tirofiban, and a score of 1 to eptifibatide in the PCI setting.

Renal dysfunction

According to data from the Italian ANMCO registry for the last few years, the incidence of renal dysfunction in Italy is ∼5% in STEMI patients and ∼10% in NSTE-ACS patients, with higher rates among patients managed conservatively than in those undergoing myocardial revascularization.^136–138 The Swedish Swedeheart registry that includes >57 000 ACS patients enrolled between 2003 and 2006, and where renal function is estimated using the modification of diet in renal disease (MDRD) formula, has shown a steady increase in the incidence of renal failure as the age increases, with a rate of 48% in the 70- to 80-year-old age group for moderate renal dysfunction (estimated GFR between 60 and 30 mL/min/1.73 m²) and 10% for severe renal dysfunction (estimated GFR < 30 mL/min/1.73 m² or patients receiving dialysis).¹³⁹

The short- and long-term prognostic impact of renal dysfunction in ACS patients is remarkable in terms of mortality, ischaemic risk and bleeding risk, and increases exponentially in patients with a glomerular filtration rate ≤30 mL/min/1.73 m²,¹⁴⁰ who are usually excluded from large clinical trials and therefore, according to the panel, deserve a separate discussion and score. Moreover, patients with renal dysfunction undergoing PCI show a 3.5-fold higher risk of definite or probable stent thrombosis compared with normal; this is the second highest risk after that associated with early discontinuation of dual antiplatelet therapy.¹⁴¹

It should be considered that the majority of antiplatelet therapies currently available show a predominantly renal route of elimination (only ticagrelor and abciximab have different metabolic elimination routes),^142,143 and this may remarkably affect the bleeding risk profile of these agents in patients with renal dysfunction.

Aspirin reduces in-hospital mortality by 64–80% in all creatinine clearance quartiles^144,145 and is associated with a reduction in myocardial infarction of ∼50% in patients with a glomerular filtration rate <60 mL/min.¹⁴⁶ In addition, some studies have shown that low doses of ASA (100 mg) in patients with chronic renal failure do not cause an increase in major bleeding events or a progression in the severity of renal failure.¹⁴⁷ Therefore, based on these efficacy and safety data, the panel assigned a score of 3 to ASA in patients with ACS and renal dysfunction.

There are limited data supporting the use of clopidogrel in this population of patients. Some sub-analyses of the CREDO¹⁴⁸ and CURE¹⁴⁹ trials have essentially reported a reduction in the efficacy of clopidogrel vs. placebo in patients with a glomerular filtration rate <60 mL/min, with a concurrent increase in the risk of transfusions and minor bleeding events. However, in consideration of the large clinical experience with clopidogrel in this population at high ischaemic and bleeding risk, the panel assigned a score of 3 to clopidogrel use in patients with a glomerular filtration rate between 59 and 31 mL/min, and a score of 1 in patients with severe renal dysfunction (glomerular filtration rate <30 mL/min).

The experience reported to date in the subgroup of renal failure patients treated with prasugrel is limited.^150,151 Unpublished data from the TRITON-TIMI 38, mainly used for a validation required by drug regulatory authorities, have not shown any differences between patients with or without renal failure who were treated with prasugrel vs. clopidogrel (HR 0.86; 95% CI 0.66–1.11; P = 0.52) (data on file).

On the contrary, a pre-specified sub-analysis of the PLATO trial in 21% of the overall population (∼3200 patients), who had a glomerular filtration rate between 30 and 60 mL/min (estimated by the Cockcroft-Gault equation), confirmed an advantage of ticagrelor vs. clopidogrel in the reduction of the primary endpoint (17.3 vs. 22.0%; HR 0.77; 95% CI 0.65–0.90), with an absolute risk reduction greater than that of patients with normal renal function (7.9 vs. 8.9%; HR 0.90; 95% CI 0.79–1.02), and an incremental advantage also in terms of all-cause mortality (10.0 vs. 14.0%; HR 0.72; 95% CI 0.58–0.89) with worsening renal function.¹⁵² This observation was associated with an increase in non-CABG-related major bleeding events of 22 and 26% according to the PLATO and TIMI classifications, respectively.¹⁵² Despite such promising findings with ticagrelor in the population with ACS and renal dysfunction, it should be considered that the observed benefit is remarkably reduced when patients are classified using the MDRD formula and that, if we take into consideration only the PLATO population without renal dysfunction, the reduction of mortality in the ticagrelor arm would drop to a non-significant 9% (as if the entire mortality benefit or the supposed efficacy caused by ticagrelor-induced inhibition of adenosine reuptake was an exclusive benefit of patients with renal dysfunction).¹⁵³For these reasons, the panel assigned a score of 2 to prasugrel in clopidogrel-naive patients who are scheduled for a PCI with the known limitations of use and with a glomerular filtration rate between 59 and 31 mL/min, and a score of 1 in patients with severe renal dysfunction (glomerular filtration rate <30 mL/min), while assigned a score of 2 to ticagrelor in all groups of patients with renal dysfunction and ACS.

As regards GPIIb/IIIa inhibitors, too, no trials have been conducted to date to specifically evaluate the use of these compounds in patients with renal dysfunction. The available sub-analyses show that the benefit produced by GPIIb/IIIa inhibitors does not appear to be influenced by the presence of renal failure, with a parallel increase in the incidence of minor and major bleedings.¹⁵⁴ Obviously, when using GPIIb/IIIa inhibitors that are eliminated by renal route (tirofiban and eptifibatide), dose reduction is recommended in these patients to avoid overdose and associated bleeding events. For tirofiban, if the estimated glomerular filtration rate is <30 mL/min/1.73 m², a 0.2 μg/kg/min bolus should be administered followed by a 0.05 μg/kg/min infusion, while for eptifibatide—that is contraindicated in patients receiving dialysis—administration of a 180 μg/kg/min bolus followed by a 1 μg/kg/min infusion is recommended when the estimated glomerular filtration rate is <50 mL/min/1.73 m². Therefore, also in consideration of the experience in patients with renal dysfunction and ACS, the panel assigned a score of 2 to the class of GPIIb/IIIa inhibitors when the glomerular filtration rate is between 59 and 31 mL/min, and a score of 1 in patients with severe renal dysfunction (glomerular filtration rate <30 mL/min).

Diabetes mellitus

It is well known that patients with ACS and hyperglycaemia or a history of diabetes mellitus have higher mortality rates when compared with patients without diabetes.^155,156 This can be partially explained by the fact that glucose metabolism changes, insufficient action of insulin, multiple metabolic conditions, and additional cellular abnormalities (e.g. increased platelet turnover, up-regulation of the P2Y12 signal, and oxidative stress) that are often associated with diabetes mellitus, lead to increased platelet aggregation.¹⁵⁷ Diabetes mellitus is also associated with a resistance to antiplatelet aggregation drugs at least four to five times higher than that observed in patients without diabetes.^158,159 As a consequence, the efficacy of antiplatelet drugs like clopidogrel in patients with diabetes, where specific studies are lacking, is strongly reduced^158,159 and this affects all-cause and cardiovascular mortality in these patients; for them, a more aggressive antithrombotic strategy—or at least a different therapy when compared with non-diabetic patients—should be considered, even assuming the twice daily administration of mild antiplatelet treatments like ASA.^158,159For this reason and considering the data successively reported of clinical superiority without any increase in the risk of bleeding of the new thienopyridines, the panel assigned a score of 1 to clopidogrel, always associated with ASA, to which the panel assigned a score of 3. Moreover, the panel suggested not to consider the results of platelet aggregation tests,^160–162 especially in this population of patients, when routinely selecting an antiplatelet therapy. From a clinical point of view and considering the benefits reported below, if a diabetic patient has ACS and is treated with clopidogrel, switching to a new thienopyridine should be considered.

To date, there is only a pharmacodynamic randomized study directly comparing the two novel P2Y₁₂ inhibitors suggesting that ticagrelor is superior to prasugrel to reduce platelet reactivity in ACS patients with diabetes mellitus. However, whether the higher potency of ticagrelor could translate into a clinical benefit needs further investigations.¹⁶³

With regard to this, data from the TRITON-TIMI 38 trial suggest that prasugrel is more effective than clopidogrel in terms of primary endpoint (cardiovascular death, myocardial infarction, and stroke) in the pre-specified subgroup of ∼3000 patients with a history of diabetes mellitus (12.2 vs. 17.0%; HR 0.70; P < 0.001, P-value for the interaction = 0.09), the efficacy being mainly related to the reduction in infarction rates (8.2 vs. 13.2%; HR 0.60; P < 0.001, P-value for the interaction = 0.02).¹⁶⁴ This benefit was observed in insulin-treated diabetic patients (14.3 vs. 22.2%; HR 0.63; P = 0.009) as well as in those who were not receiving insulin (11.5 vs. 15.3%; HR 0.74; P = 0.009), with no significant increase in major bleeding according to TIMI classification. Moreover, prasugrel significantly reduced stent thrombosis in diabetic patients (3.6 vs. 2.0%; HR 0.52; P = 0.007)¹⁶⁴ and seems to be beneficial in this patient population also in reducing the frequency of recurrent clinical events following index hospitalization due to ACS, when compared with patients without a clinical history of diabetes mellitus (11.7 vs. 27%).¹⁶⁵

Of the 4662 patients with a history of diabetes mellitus enrolled in the PLATO trial, the reduction of ticagrelor vs. clopidogrel in terms of composite primary endpoint (HR 0.88; 95% CI 0.76–1.03), all-cause mortality (HR 0.82; 95% CI 0.66–1.01), and stent thrombosis (HR 0.65; 95%CI 0.36–1.17), without any increase in major bleedings (HR 0.95; IC 95% 0.81–1.12), was consistent with the overall population, with no interaction between diabetic status and treatment.¹⁶⁶ No heterogeneity was observed between patients who received insulin or not, and ticagrelor significantly reduced the incidence of primary endpoint, all-cause mortality, and stent thrombosis in patients with a glycated haemoglobin level higher than the median value (HR 0.80, 95% CI 0.70–0.91; HR 0.78, 95% CI 0.65–0.93; and HR 0.62, 95% CI 0.39–1.00, respectively) with a similar frequency of bleeding (HR 0.98; 95% CI 0.86–1.12).¹⁶⁶Taking into consideration the data reported in the various sub-analyses of the key studies (TRITON-TIMI 38 and PLATO), the panel assigned a score of 3 to prasugrel (in clopidogrel-naive patients, when there are no contraindications to the use of this agent, when coronary anatomy is known and a PCI is planned) and a score of 2 to ticagrelor in ACS patients with a history of diabetes mellitus who are scheduled to undergo whatever therapeutic strategy.

Data on GPIIb/IIIa inhibitors show a particular benefit of these agents in this ACS high-risk patient population, especially if undergoing a PCI,¹⁶⁷ but they date back to a period when oral antiplatelet therapy was mild.¹⁶⁸ This is why it is necessary to re-consider the use of GPIIb/IIIa inhibitors in diabetic patients with the new thienopyridines currently on the market. In the light of the considerable benefit of GPIIb/IIIa inhibitors in the diabetic population reported in literature, the panel assigned a score of 3 to this class of drugs, reminding that data on the efficacy/safety ratio of the association between GPIIb/IIIa inhibitors and the new oral inhibitors of platelet receptor P2Y₁₂ is scanty.

Elderly patients

Elderly patients, often excluded from clinical trials, have specific issues to be addressed, including (i) reduced lean body mass that can vary the distribution volume of hydrophilic and lipophilic drugs, (ii) slow bowel transit time that can modify drug absorption, (iii) polypharmacy that can raise issues of drug–drug interactions, and (iv) comorbidities, i.e. renal insufficiency that can lead to accumulation of drugs that are eliminated via renal route.¹⁶⁹

The meta-analysis performed by the Antithrombotic Trialists Collaboration suggests that ASA, when given for secondary prevention, reduces the composite risk of mortality for cardiovascular causes, myocardial infarction, and stroke by 23% in patients aged over 65.¹¹⁵ Elderly patients should receive ASA at the minimum effective dose (75–150 mg), even if this dose is also associated with a higher risk of gastrointestinal bleeding, in particular in patients with a history of bleeding.¹⁷⁰Scientific robustness as well as clinical experience with ASA in ACS elderly patients was deemed sufficient to assign a score of 3 to this treatment.

As regards clopidogrel, most of the data come from the CURE study, when compared with younger patients, the RR reduction for the incidence of the primary endpoint (vascular death, myocardial infarction, and stroke) was lower in elderly patients (13 vs. 29%), although it translated into a similar reduction in the absolute risk (2.0 vs. 2.2%), probably due to the greater prevalence of vascular and coronary diseases with aging.⁹ Moreover, it should be noted that several observational and pharmacodynamics studies agree in suggesting that, in elderly patients, high on-clopidogrel platelet reactivity increases probably due to the higher number of comorbidities that are more specifically associated with this type of issues, e.g. diabetes mellitus and renal failure.¹⁷¹Based on these data, on the large experience accumulated over the years as well as on the relative safety of this agent, the panel assigned a score of 3 to clopidogrel in ACS elderly patients.

The comparison of the efficacy of the new oral inhibitors of platelet receptor P2Y₁₂ in the elderly ACS population shows overlapping percentages between the new molecules (prasugrel and ticagrelor) and clopidogrel (18.3 vs. 17.2%), notwithstanding the differences between the key trials (TRITON-TIMI 38 and PLATO) that were previously discussed^41,172 (Figure 5). Therefore, prasugrel and ticagrelor, at least from a numerical point of view, result into a mild reduction in the ischaemic endpoint in the elderly population, but without achieving statistical significance since the numbers were low and the power of the study was obviously calculated on the primary endpoint of the overall population. Conversely, non-CABG-related major bleeding was slightly increased in both treatment arms of the new antiplatelet drugs vs. clopidogrel. It is well known that in the TRITON-TIMI 38 trial, patients aged ≥75 who were treated with prasugrel had an increased risk of developing major and fatal bleeding vs. clopidogrel (4.2 vs. 3.4%; HR 1.36; P = 0.24), resulting into a neutral net clinical benefit (21.7 vs. 21.5%; HR 0.99; P = 0.92).⁴¹ With regard to this, the FDA declared that a 10 mg prasugrel dose in elderly patients undergoing percutaneous revascularization can be considered if high-risk factors are present, including diabetes or prior myocardial infarction;¹⁷³ conversely, European Medicines Agency suggested a reduced dose of 5 mg in all ACS patients aged over 75 planned to undergo PCI, in the absence of wide clinical safety data on this dosage but only based on pharmacokinetics studies.^174–176 In this respect, the cross-over study called GENERATIONS (Comparison of Prasugrel and Clopidogrel in Very Elderly and Non-Elderly Patients With Stable Coronary Artery Disease), carried out in 73 elderly patients (mean age 79 ± 3 years) with stable angina, showed that a 5 mg dose of prasugrel is not inferior to the 10 mg dose with a better pharmacodynamics response and a lower number of non-responders vs. clopidogrel 75 mg.¹⁷⁵ In the recent TRILOGY ACS trial conducted in ∼2000 ACS patients aged ≥75 undergoing conservative treatment, no differences were observed in terms of efficacy (composite primary endpoint of cardiovascular death, myocardial infarction, and stroke) between 5 mg prasugrel and 75 mg clopidogrel (HR 1.03; 95% CI 0.86–1.22), with a similar incidence of major bleeding according to the TIMI classification (4.1 vs. 3.4%; HR 1.09; 95% CI 0.57–2.08)¹¹⁰ (Figure 6). The use of the lower dose seems to be reasonable also in patients with an additional contraindication to the use of prasugrel, i.e. patients with a body weight <60 kg, as suggested by the cross-over study on stable patients called FEATHER (Comparison of Prasugrel and Clopidogrel in Low Body Weight vs. Higher Body Weight with Coronary Artery Disease).¹⁷⁷ To establish if a 5-mg dose is superior to a standard dose of clopidogrel in elderly patients undergoing myocardial revascularization is the objective of the Italian ongoing ELDERLY-ACS 2 trial, where ∼2000 ACS patients aged ≥75 undergoing PCI will be randomized to receive clopidogrel 300 mg loading dose followed by 75 mg maintenance dose and prasugrel 60 mg loading dose followed by 5 mg maintenance dose with a 12-month follow-up for the evaluation of a composite primary endpoint of death, myocardial infarction, stroke, re-hospitalization for cardiovascular causes, and severe bleeding according to the Bleeding Academic Research Consortium classification. Looking forward to new data and based on the information available, the panel assigned a score of 1 to prasugrel in the elderly patient subgroup.

Figure 5

Incidence of primary endpoint and major bleeding events in the elderly population (≥75 years) included in the key clinical trials on new oral inhibitors of platelet receptor P2Y₁₂. CABG, coronary artery bypass graft; Adj HR, adjusted hazard ratio; CI, confidence interval.

Open in new tabDownload slide

Figure 6

Composite and individual efficacy and safety endpoints in the elderly population (≥75 years) enrolled in the Targeted Platelet Inhibition to Clarify the Optimal Strategy to Medically Manage Acute Coronary Syndromes trial.

Open in new tabDownload slide

As for ticagrelor, the pre-specified subgroup of PLATO patients aged over 75 (n = 2878) was evaluated for several safety endpoints (major bleeding, dyspnoea, and ventricular pauses) and efficacy endpoints (composite endpoint, death from cardiovascular causes, death from any causes, myocardial infarction, stroke, and stent thrombosis) and all interactions were negative; therefore, the results from the main trial do not depend on the older age in general (≥75 years), nor on age at 5-year intervals (increasing from 25 to 95 years).¹⁷²In the light of this encouraging data, the panel assigned a score of 2 to ticagrelor while awaiting new data and additional clinical experience on the safety profile of this new agent in the ACS elderly population.

Finally, GPIIb/IIIa inhibitors were put together in the same category, taking into consideration a few post hoc analyses of randomized clinical trials where elderly patients were not excluded, e.g. the Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications study,¹⁷⁸ and the meta-analysis by Boersma et al.¹²¹ where the beneficial delta of their administration declined with aging to reach 4% in patients over 70 years, thus increasing the risk of major bleeding by 62%. These results on the use of GPIIb/IIIa inhibitors in elderly patients may be improved by a very careful patient selection, e.g. by excluding low-risk patients, patients undergoing conservative strategy, or patients with serious co-morbidities, such as renal failure, in which dose adjustment is absolutely recommended in case eptifibatide and tirofiban are used in order to avoid overdose (see section on renal failure). Therefore, from the evaluation of the scarce available data and the RR–benefit ratio seen in elderly ACS patients overall, the panel assigned a score of 1 to the GPIIb/IIIa inhibitor class.

High risk of bleeding

Over the last few years, great attention has been given to the incidence, and related clinical impact, of bleeding events induced by the use of increasingly effective antithrombotic treatments, and some randomized studies have included major bleeding events as safety and primary efficacy endpoints when evaluating new drugs in ACS patients, together with the classical incidence of mortality, myocardial infarction, stroke, and urgent revascularization.¹⁷⁹

It is well known that guidelines recommend early risk stratification in ACS patients.⁵⁵ This risk stratification includes an evaluation of the ischaemic risk (preferably based on GRACE score) as well as of the risk of bleeding events (to guide the selection of the therapeutic and anticoagulant strategies).⁵⁵ The score used to assess bleeding risk in the Can Rapid risk stratification of Unstable angina patients Suppress Adverse outcomes with Early implementation of the ACC/AHA guidelines (CRUSADE) study is derived from the ACS registry of the same name and consists of a score to assess the risk of major bleeding events based on eight parameters: low haematocrit level, low creatinine clearance level, female gender, presence of diabetes mellitus, heart failure signs, low blood pressure (or excessively high blood pressure), high heart rate, and presence of other vasculopathies.¹⁸⁰ A score of <21 identifies very low risk patients, from 21 to 30 refers to low-risk patients, from 31 to 40 the risk is moderate, from 41 to 50 the risk is high, while a score of >50 identifies a very high risk.¹⁸⁰ This stratification of the risk of bleeding events seems crucially important considering that there is a close relationship between how aggressive an antithrombotic therapy is and the incidence of bleeding events; therefore, clinicians should be able to identify the ideal therapeutic window in which the risk of ischaemic and bleeding events is minimized. Bleeding can cause shock, anaemia, need for transfusion that concur to the activation of inflammation and thrombotic mechanisms, but also to the reduction or discontinuation of recommended antithrombotic treatments and this can result into higher long-term mortality rates. In ACS, the influencing factors for the bleeding risk are related to patient characteristics (e.g. body weight, renal function, and comorbidities), to the drug (drug potency, therapeutic window, and therapy duration) as well as to the quality of treatment (number of antithrombotic drugs used, combinations of drugs, possible invasive strategy, vascular access, lack of use of gastroprotective drugs, treatment adherence, and complication management).^181,182

Associazione Nazionale Medici Cardiologi Ospedalieri has recently completed the management of patients with ACS in the real-world practice in Italy: an outcomes research focused on the use of ANTithRombotic Agents registry¹³⁸ on 6400 ACS patients (of which 45% with STEMI) hospitalized from April 2009 and December 2010 in 52 Italian intensive cardiac care units. This registry was designed to study the real-world incidence of bleeding events in ACS patients hospitalized in centres having a high adherence to guideline recommendations.

The in-hospital incidence of bleeding events was 1.2% (1.4% in STEMI and 1.1% in NSTE-ACS), much lower than the incidence reported in other registries,¹⁸³ with a linear correlation between major bleeding and 6-month mortality rates.¹³⁸

This frail population has been generally excluded from the wide clinical trials and due to the scarce available data, the absolute lack of direct or indirect comparison data between the different antiplatelet drugs, the fact that in the key trials (PLATO and TRITON-TIMI 38) the new inhibitors of platelet receptor P2Y₁₂ were associated with increased non-CABG-related major bleeding events in a selected population from which patients at high risk of bleeding events were excluded, and based on the large clinical practice of the recent years, the panel assigned a score of 3 to the use of clopidogrel in patients with high risk of bleeding (CRUSADE score ≥50).

In the light of clinical reasonability and recent clinical trials, e.g. the Prolonging Dual Antiplatelet Treatment After Grading Stent-Induced Intimal Hyperplasia Study study,¹⁸⁴ which documented the 2-year safety of a dual antiplatelet strategy with ASA and clopidogrel for 6 months only vs. 24 months in terms of death, myocardial infarction, and stroke (10.1 vs. 10.0%; HR 0.98; 95% CI 0.74–1.29; P = 0.91) with a lower incidence of major bleeding events in the group with lower duration of the dual antiplatelet treatment, in an all-comers population of over 2000 patients (of which ∼75% of patients had ACS and 33% had STEMI) undergoing PCI, the panel assigned a score of 3 to the ASA + clopidogrel association for 6 months in this population at high risk of developing bleeding events.

Acute coronary syndrome patients who receive oral anticoagulant treatment (OAT) account for ∼5–6% of the total population. Oral anticoagulant treatment per se increases the risk of bleeding events and the addition of ASA alone, as well as the triple therapy (OAT + ASA + thienopyridine) that has been widely used so far,¹⁸⁵ exponentially increases the incidence of bleeding events. In the recent What is the Optimal antiplatElet & Anticoagulant Therapy in Patients With Oral Anticoagulation and Coronary StenTing trial, 573 patients treated with OAT undergoing PCI were openly randomized to receive clopidogrel alone or clopidogrel + ASA, showing a greater safety of the dual therapy vs. triple therapy in terms of bleeding events, mostly of minor extent (HR 0.36; 95% CI 0.26–0.50, P < 0.0001).¹⁸⁶ No data are available on the association of OAT and ASA and new inhibitors of platelet receptor P2Y₁₂, but for a small study on 377 patients undergoing PCI and treated with OAT and ASA for 6 months in association with clopidogrel or prasugrel, and the latter treatment increased the risk of major bleeding events of approximately four times when compared with the standard treatment.¹⁸⁷Based on these data, the panel deemed it reasonable to assign a score of 3 to the addition of clopidogrel in patients with ACS who need to receive old or new oral anticoagulants. Finally, the panel suggested that, since no data are available on the risk of bleeding events with triple therapy with the new P2Y₁₂ inhibitors, in case patients develop conditions that were not present at hospitalization following ACS for which OAT is needed (e.g. atrial fibrillation) and they are already being treated with ASA + prasugrel or ticagrelor, it is reasonable to discontinue the new antiplatelet drug and to switch to clopidogrel.

The multi-centre randomized PIONEER AF-PCI study will compare the different treatment strategies in ∼2100 patients with atrial fibrillation undergoing PCI: (i) rivaroxaban 15 mg/day in addition to an inhibitor of platelet receptor P2Y₁₂ (clopidogrel 75 mg/day or prasugrel 10 mg/day or ticagrelor 180 mg bid); (ii) rivaroxaban 2.5 mg bid, ASA 75–100 mg/day in addition to an inhibitor of platelet receptor P2Y₁₂ (clopidogrel 75 mg/day or prasugrel 10 mg/day or ticagrelor 180 mg bid); (iii) OAT (INR 2.0–3.0), ASA 75–100 mg/day in addition to an inhibitor of platelet receptor P2Y₁₂ (clopidogrel 75 mg/day or prasugrel 10 mg/day or ticagrelor 180 mg bid).¹⁸⁸ The primary endpoint will be safety and will include major bleeding events, major and minor bleeding events according to TIMI classification and clinically relevant bleeding events.¹⁸⁸

Patients treated with clopidogrel (switching)

During the diagnosis and treatment approach of ACS patients, it is possible to combine a number of different antithrombotic drugs and to switch from one drug to another within the same class of drugs, thus obtaining a thousand possible combinations.¹⁸⁹

Within this context, we should consider the switch from clopidogrel to the new oral antiplatelet drugs. Few data are available on this and mostly based on pharmacokinetics studies of small sample size. Hence, we often have data on the antiplatelet performance of a pharmacological switching strategy without having robust data on the related clinical efficacy or safety profile. A recent study with a clinical and non-mechanistic endpoint, in which 606 ACS patients were randomized to clopidogrel followed by prasugrel (both administered at a loading dose) vs. prasugrel alone at a loading dose, suggested that switching is associated with a good safety profile in terms of bleeding events, but also with an increase in the number of cardiac adverse events, essentially due to a higher incidence of urgent CABG.¹⁹⁰ Recently, the GReek AntiPlatelet REgistry Registry showed that in-hospital switching represents common clinical practice, occurring in 35.5% of ACS patients undergoing PCI.¹⁹¹ Clinical factors and regional practice differences seem to affect this strategy's choice, while switching to a novel agent may be associated with higher risk of bleeding.¹⁹¹

There are two types of switching: (i) the so-called ‘cross-over’, in which each patient is a control for himself and initially receives drug A, then drug B and then drug A again; there are seven papers published on stable patients and four on ACS patients in this setting; (ii) the real switch from one drug to another, with three studies on healthy subjects or patients with stable coronary disease and five studies on ACS patients that have been published in this setting (eight studies with prasugrel and none with ticagrelor) (Table 5),^192–199 besides the large PLATO clinical trial, in which switching from clopidogrel to ticagrelor was allowed (and performed in 46% of cases).⁴⁸

The SWitching Anti Platelet (SWAP) study is a Phase 2 trial in which 140 ACS patients treated with clopidogrel were randomized to three treatment arms: clopidogrel 75 mg, prasugrel 10 mg, or prasugrel at a loading dose (60 mg) followed by maintenance treatment (10 mg), that essentially showed that a rapid and effective antiplatelet effect during ACS (in terms of platelet reactivity units) is obtained by administering a loading dose of prasugrel.¹⁹³ Recently, in the Transferring from Clopidogrel Loading Dose to Prasugrel Loading Dose in Acute Coronary Syndrome Patients (TRIPLET) study, 282 ACS patients undergoing PCI were randomized to three strategies of loading dose (always with prasugrel in the cath lab after coronary angiography): placebo and prasugrel 60 mg, clopidogrel 600 mg followed by prasugrel 60 mg, and clopidogrel 600 mg followed by prasugrel 30 mg, showing that prasugrel loading dose administered in the cardiac cath lab was not inferior, in terms of platelet reactivity units measured 6 h after PCI, to the treatment arms where patients, besides the loading dose received in the cath lab, had also received 600 mg clopidogrel as pre-treatment.¹⁹⁹

Only one study performed a head-to-head comparison of prasugrel and ticagrelor (treatments were also changed with cross-over design after 15 days of treatment) in 44 ACS patients undergoing PCI and non-responder to clopidogrel 24 h after the procedure, suggesting a higher platelet inhibition of ticagrelor when compared with prasugrel (32.9 vs. 101.3 platelet reactivity units; P < 0.001).²⁰⁰ Notwithstanding the huge amount of pharmacodynamics studies and the biological plausibility of switching, especially between molecules with the same active metabolite (clopidogrel/prasugrel), the panel deemed it desirable to have more clinical experience as well as safety data on this therapeutic strategy before issuing any recommendations on its use. Moreover, as of today, switching in patients with a high risk of bleeding events as well as planned switch (identification of patient categories to which clopidogrel should be initially administered and the antiplatelet therapy should be changed in case of PCI) are not advisable strategies. In the light of the reported evidence and clinical impact of published data, the panel assigned a score of 2 to clopidogrel/prasugrel switch and a score of 3 to clopidogrel/ticagrelor switch.

Surgery candidates

The fear that coronary disease in ACS patients may expand to such an extent that requires a CABG affects the therapeutic choices and strategies, often leading to an under-treatment in the early phases of ACS in view of a possible intervention, surgical urgent revascularization, that with the increasing experience of the operators and interventional technology is becoming increasingly rare.^114,201 Moreover, the recent surgical technologies, i.e. off-pump surgery, emerging blood salvaging, and platelet transfusion techniques, have considerably reduced the risk of major bleeding events in ACS patients undergoing surgery while treated with a dual antiplatelet therapy.²⁰²

However, a number of studies have shown that the pre-surgical use of ASA and clopidogrel is associated with an increased risk of bleeding, transfusions, and surgical re-intervention due to haemorrhage.^203–209 In a study carried out in 14 centres on 350 patients undergoing CABG, the risk of surgical re-intervention due to haemorrhage increased by three times in patients who had received clopidogrel within 5 days before surgery.²¹⁰

This time interval associated with an increased risk of major bleeding in patients treated with clopidogrel undergoing CABG is consistent with the one reported in a sub-analysis of the CURE trial.¹⁰⁸ Hence, the recommendation that clopidogrel-treated patients for whom a CABG has been scheduled should discontinue the drug 5–7 days before surgery,⁵⁵ independently of the values of platelet reactivity monitoring.²¹¹

As regards prasugrel, safety data in CABG are relatively few, since they can be only obtained from the TRITON-TIMI 38 population (a study essentially designed to address included patients to PCI) where only 1% of the population enrolled underwent surgical revascularization.²¹² In this small subgroup of patients, prasugrel was associated with a reduction in all-cause mortality (OR 0.26; P = 0.025), at the expense of an increase in major bleeding events in 13% of cases vs. 3% of patients treated with clopidogrel (HR 4.73; 95% CI 1.90–11.82; P < 0.001),²¹² which means that the practical behaviour in case of early prasugrel administration in patients undergoing CABG is to discontinue the treatment at least 7 days before surgery.

In the PLATO study, ∼2000 patients underwent CABG after randomization. The study protocol provided for discontinuation of ticagrelor 24–72 h and clopidogrel 5 days before surgery.²¹³ In general, near 1200 patients underwent CABG within 7 days from study drug discontinuation. In this PLATO subgroup, ticagrelor, with a median discontinuation time of 4 days, showed an improvement of the primary endpoint that was consistent with the global study results, with no difference in CABG-related bleeding events in the two treatment arms, but with a reduction in the total mortality rate ranging from 9.7% for clopidogrel to 4.7% for ticagrelor (HR 0.49; 95% CI 0.32–0.77; P < 0.01) and in cardiovascular mortality ranging from 7.9 to 4.1% (HR 0.52; 95% CI 0.32–0.85; P < 0.01).²¹³ This reduction in mortality rates seems to be attributable to ticagrelor favourable pharmacokinetic properties (reversibility with on/off effect), that in this context translated into a sharp reduction in major bleeding events and perioperative infections vs. clopidogrel.²¹⁴ The European guidelines recommend ticagrelor discontinuation before undergoing a surgery within 5 days.⁵⁵

With regard to discontinuation strategies of antiplatelet drugs based on the thrombotic and haemorrhagic risks in stable patients carrying a coronary stent who require surgery, refer to the previously published GISE-ANMCO consensus.²¹⁵ As regards ACS patients carrying a coronary stent who need to undergo urgent surgery, it is clear that, following the preventive discontinuation of oral antiplatelet drugs according to the above time schedule, bridge therapy protocols with short plasma half-life intravenous drugs (i.e. GPIIb/IIIa inhibitors) are very important. For example, GPIIb/IIIa inhibitors were used as bridge therapy in a protocol proposed for an observational and perspective registry²¹⁶ in which clopidogrel should be discontinued 5 days before surgery (as suggested by the guidelines) and tirofiban should be started (0.4 mg/kg/min bolus followed by infusion of 0.1 mg/kg/min with the exception of patients with a glomerular filtrate rate ≤30 mL/min in which the maintenance dose should be reduced by half) starting from day −3 until 4 h before surgery (8 h before surgery in patients with severe renal failure), and the dual oral antiplatelet therapy should be restarted as soon as possible after surgery. The panel deemed that the same time schedule (starting the GPIIb/IIIa inhibitor 3 days before surgery) can be used also in case of discontinuation of the new inhibitors of platelet receptor P2Y₁₂. With regard to this, as of today there are no known data or protocols on the use of abciximab in bridge therapy; therefore, the panel assigned a score of 0, while published data are available, though only limited, to support the use of tirofiban,^184–216to which the panel assigned a score of 2; more limited evidence is available on eptifibatide,^217,218even if this molecule is similar to tirofiban and is widely used, to which the panel assigned a score of 1. The panel noted that the use of bridge therapy is outside all guidelines and the drugs are used off-label; therefore, patients should give their informed consent for the compassionate use of the drug. Moreover, it should be highlighted that GPIIb/IIIa inhibitors need 4–6 h from discontinuation to take platelet function values back to their baseline values and that prolonged infusions could increase the risk of bleeding events.

With regard to this, cangrelor, thanks to its pharmacokinetic properties, seems to be an absolutely promising drug in bridge therapy. The BRIDGE (Maintenance of Platelet Inhibition with Cangrelor) trial randomized 210 patients with ACS or stent waiting to undergo CABG, to receive cangrelor 0.75 μg/kg or placebo for 2–7 days until 1–6 h before surgery.²¹⁹ Patients randomized to cangrelor had lower platelet reactivity when compared with placebo (primary endpoint) with an equal incidence of major bleeding events, according to the GUSTO classification (11.8 vs. 10.4%, RR 1.1; 95% CI 0.5–2.5; P = 0.7).²¹⁹

Finally, it is important to highlight that in case of urgent CABG and major bleeding events, also during a major surgery, in patients treated with oral antiplatelet therapy or GPIIb/IIIa inhibitors, adequate measures to ensure haemostasis should be complied with, in the first place platelet transfusion.²²⁰

As regards ticagrelor, there are no data on the haemostatic benefit of platelet transfusion since circulating ticagrelor can inhibit transfused platelets, even if, in case of massive bleeding, this seems to be the only possible therapeutic strategy, trying to infuse a higher number of platelet bags than normal in order to elute the drug effect. The antifibrinolytic therapy (aminocaproic acid or tranexamic acid) and/or recombinant factor VIIa could increase haemostasis. In any case, antiplatelet treatments should be reintroduced 24 h after resolution of the major bleeding event.

Conflict of interest: L.D.L. has received honoraria for lectures at symposia and advisory boards from Abbott Vascular, AstraZeneca, Boehringer Ingelheim, Daiichi Sankyo, Lilly, Menarini, and The Medicines Company. L.B. has received honoraria for lectures at symposia and advisory boards from AstraZeneca, Daiichi Sankyo, Lilly, Menarini, and The Medicines Company. M.V. has received honoraria for lectures at symposia and advisory boards, and/or research grants from Abbott Vascular, AstraZeneca, CID, Cordis, Correvio, Daiichi Sankyo, Lilly, Medtronic, Merck, Sanofi-Aventis, St. Jude, Terumo, and The Medicines Company.

Appendix

Subgroup coordinators

1. Non-ST elevation acute coronary syndrome: Francesco Prati (Rome), Marco Valgimigli (Rotterdam).

2. ST elevation acute myocardial infarction: Leonardo Bolognese (Arezzo), Giuseppe De Luca (Novara).

3. Conservative treatment: Mario Marzilli (Pisa), Zoran Olivari (Treviso).

4. Coronary angiography: Leonardo De Luca (Rome), Stefano De Servi (Legnano).

5. Renal dysfunction: Gianni Casella (Bologna), Paolo Sbarzaglia (Cotignola).

6. Diabetes mellitus: Claudio Cavallini (Perugia), Marco Tubaro (Roma).

7. Elderly patients: Davide Capodanno (Catania), Stefano Savonitto (Reggio Emilia).

8. High risk of bleeding: Francesco Maria Bovenzi (Lucca), Giuseppe Di Pasquale (Bologna).

9. Switching from clopidogrel: Guido Parodi (Florence), Gennaro Sardella (Rome).

10. Surgery candidates: Maddalena Lettino (Rozzano), Luigi Oltrona Visconti (Pavia).

Working group

R. Abbate (Florence), F. Andreotti (Rome), L. Bolognese (Arezzo), G. Biondi-Zoccai (Rome), F.M. Bovenzi (Lucca), D. Capodanno (Catania), R. Caporale (Cosenza), P. Capranzano (Catania), N. Carrabba (Florence), G. Casella (Bologna), C. Cavallini (Perugia), R. Ceravolo (Catanzaro), P. Colombo (Milan), M.R. Conte (Turin), S. Cordone (Savona), A. Cremonesi (Cotignola), G.B. Danzi (Milan), M. Del Pinto (Perugia), G. De Luca (Novara), L. De Luca (Rome), S. De Servi (Legnano), E. Di Lorenzo (Avellino), G. Di Pasquale (Bologna), G. Esposito (Naples), R. Farina (Salerno), A. Fiscella (Catania), D. Formigli (Benevento), S. Galli (Milan), P. Giudice (Salerno), G. Gonzi (Parma), C. Greco (Rome), N.B. Grieco (Milan), L. La Vecchia (Vicenza), M. Lazzari (Lucca), C. Lettieri (Mantova), M. Lettino (Rozzano), U. Limbruno (Grosseto), A. Lupi (Novara), A. Macchi (Busto Arsizio), M. Marini (Ancona), M. Marzilli (Pisa), A. Montinaro (Lecce), G. Musumeci (Bergamo), A. Navazio (Guastalla), Z. Olivari (Treviso), L. Oltrona Visconti (Pavia), J.A. Oreglia (Milan), F. Ottani (Forlì), G. Parodi (Florence), G. Pasquetto (Este), G. Patti (Rome), A. Perkan (Trieste), G.P. Perna (Ancona), E. Piccaluga (Milan), F. Piscione (Salerno), F. Prati (Roma), S. Rakar (Trieste), R. Ravasio (Milano), F. Ronco (Mestre), R. Rossini (Bergamo), A. Rubboli (Bologna), F. Saia (Bologna), G. Sardella (Rome), G. Satullo (Messina), S. Savonitto (Reggio Emilia), P. Sbarzaglia (Cotignola), G. Scorcu (Cagliari), N. Signore (Bari), G. Tarantini (Padova), P. Terrosu (Sassari), L. Testa (Milan), M. Tubaro (Rome), S. Valente (Florence), M. Valgimigli (Rotterdam), F. Varbella (Rivoli), M. Vatrano (Catanzaro).

References