Reports

This report is based on medical evidence presented at sanctioned medical congress, from peer reviewed literature or opinion provided by a qualified healthcare practitioner. The consumption of the information contained within this report is intended for qualified Canadian healthcare practitioners only.

PHYSICIAN PERSPECTIVE - Based on presentations during the 81st Scientific Sessions of the American Heart Association

New Orleans, Louisiana / November 9-12, 2008

Editorial overview:

John Eikelboom, MBBS, MSc, FRACP, FRCPA, FRCPC

Division of Haematology and Thromboembolism, Associate Professor of Medicine, McMaster University, Hamilton, Ontario

The AHA/ACC guidelines recommend ASA be used at a dose of 165 to 325 mg/day for one month following placement of a bare metal stent, for three months after the placement of a sirolimus-eluting stent and for six months after placement of a paclitaxel-eluting stent. Thereafter, patients should receive 75 to 162 mg/day. But as Dr. Peter Berger, Director, Center for Clinical Studies, Geisinger Clinic, Danville, Pennsylvania, argued, rates of cardiovascular (CV) death, myocardial infarction (MI) and stroke in CURE were actually higher in patients who received ASA >200 mg/day (13.7%) than those who received <100 mg/day (9.7%). Rates of major bleeding (4%) were essentially double those in the highest-dose ASA group compared with the lowest-dose group.

Similarly in CHARISMA, there was very little difference in efficacy between those receiving <100 or >100 mg/day. Severe bleeding rates were higher for those on the >100 mg/day dose (1.8%) compared to those on the <100 mg/day dose (1.4%). This demonstrated that antiplatelet efficacy is not reduced with lower doses of ASA, hence Dr. Berger’s recommendation to use ASA 81 mg/day in this patient population. Regarding the use of thienopyridine therapy, Dr. Berger suggested that we really do not know how long treatment should be continued after placement of a drug-eluting stent (DES).

Data from a CHARISMA subset indicated that dual antiplatelet therapy in patients with symptomatic vascular disease was associated with a lower risk of CV death, MI or stroke compared with single antiplatelet therapy and that benefit extended beyond one year. Continued use of dual antiplatelet therapy does carry a risk of bleeding, speakers agreed. Conversely, given that the true duration of risk of DES thrombosis is still unknown, nor is the optimal time during which to use antiplatelet therapy to reduce that risk, there is a certain rationale for not stopping thienopyridine therapy in high-risk patients, especially if they are at low risk for bleeding. Stopping clopidogrel five days prior to surgery lessens bleeding risk, as was shown in CURE patients receiving both a thienopyridine and ASA.

Nevertheless, as pointed out by Dr. Philippe Steg, INSERM, Université Paris VII–Denis Diderot, France, withdrawal of antiplatelet therapy prior to surgery exposes patients to a protracted risk of thrombosis during the perioperative interval. The dangers of premature discontinuation have also been documented. In the PREMIER Registry, for example, investigators reported that premature discontinuation of thienopyridine therapy after placement of a DES was significantly associated with cardiac rehospitalization as well as mortality at 12 months’ follow-up. A sharp increase in all-cause mortality and MI was also seen in the Veteran’s Administration Hospital’s ACS Registry in the days and weeks following thienopyridine withdrawal, not only in patients treated with PCI but even among patients who were treated medically. Among 1358 patients in the Paris Registry admitted for ACS, over 68% were not taking oral antiplatelet agents at the time of admission. Over 26% were prior uses, but of those who were admitted for ACS, 5.4% had recently had their oral antiplatelet agents withdrawn, all of which suggests that withdrawal in the setting of ACS is cause for concern.

Continuous use of oral antiplatelet agents increases bleeding risk, but not all patients are at equal risk for bleeding. Consequently, it is particularly important to try and identify which patients are at high risk for bleeding, particularly if considering long-term dual antiplatelet therapy after DES placement. (This is less problematic following placement of bare metal stents when extended duration dual antiplatelet therapy is less critical.) Using data from the REACH Registry, Dr. Steg and colleagues developed a simple risk score to predict bleeding risk in this patient group. The score is based on nine risk factors for major bleeding including age, presence of peripheral arterial disease, diabetes, heart failure, hypercholesterolemia, hypertension, smoking and use of antiplatelet and oral anticoagulant therapy. As he suggested, using this risk score prior to initiation of antiplatelet therapy may facilitate decisions about long-term antiplatelet use.

Bleeding in ACS Patients

Why major and even minor bleeds need to be avoided in ACS patients wherever possible was discussed by several different speakers here, among them Dr. Harvey White, Auckland City Hospital, New Zealand. In a meta-analysis of over 34,000 ACS patients, the association between bleeding within the first 30 days of the index event and death was striking, with mortality rates approaching 12% at 30 days among patients with a major bleed vs. approximately 2% for those without a major bleed.

The ISAR meta-analysis of 5384 patients undergoing PCI demonstrated that 14.1% of those who bled following PCI had died at one year compared with only 3.3% of those who had no bleeding. The same meta-analysis also demonstrated that any TIMI bleeding event—bleeding within 30 days, MI within 30 days or urgent revascularization within 30 days—was an independent predictor of one-year mortality, Dr. White pointed out.

Several other large randomized studies, including the ACUITY trial, also demonstrated a large impact of major bleeding on 30-day mortality rates. In addition, non-coronary artery bypass graft (CABG) major bleeding and MI at 30 days in ACUITY significantly affected mortality outcomes at one year, with mortality rates approaching 30% among patients who had both a major bleed and an MI compared to those with major bleed alone (approximately12% mortality rate), an MI alone (8.6% mortality rate), or neither an MI nor a major bleed (3.4% mortality rate). Based on ACUITY data, major bleeding could be said to affect mortality risk at one year to a greater extent than an MI, Dr. White indicated.

Dr. Daniel Simon, Chief, Cardiovascular Medicine, University Hospitals Case Medical Center, Cleveland, Ohio, acknowledged that bleeding is associated with mortality but he did not feel that the association between the two was necessarily straightforward. In the GRACE Registry, for example, bleeding was associated with a greater likelihood of in-hospital mortality but not mortality after hospital discharge, he pointed out. Other factors also contribute to short-term mortality risk including creatinine kinase-myocardial band (CK-MB) levels. Patients with levels >10 times the ULN in-hospital had almost a 20% mortality rate at six months compared to those whose levels were normal, among whom mortality rate was <5%. There is also a clear association between in-hospital mortality and discontinuation of any antithrombotic agent, Dr. Simon noted. His interpretation of the ACUITY trial was that there were multiple predictors of mortality (although the odds ratio for major bleeding was higher than for any other predictor). Some PCI trials (e.g. HORIZONS) in STEMI patients also demonstrated a higher hazard ratio (HR) for mortality with reinfarction (9.75) than with major bleeding (HR=4.66), Dr. Simon pointed out.

His own interpretation of the ISAR group of studies was that bleeding, MI and urgent target vessel revascularization at 30 days had comparable power for predicting one-year mortality at a HR of 2.96, 2.29 and 2.49, respectively. Thus, he argued, bleeding is complex and associated with a number of significant comorbidities that in themselves predict death. Nevertheless, he confirmed that bleeding is dangerous to the patient and costly for healthcare and should be avoided.

Many Different Definitions

Importantly, as pointed out by Dr. Shamir Mehta, Associate Professor of Medicine, McMaster University, Hamilton, Ontario, physicians need to appreciate that there are many different definitions of major bleeding that are used in clinical trials and not all carry the same prognostic weight. The challenge therefore is to understand what kinds of bleeds are prognostic and which ones are not, especially when comparing bleeding rates across antiplatelet trials.

In the CURE trial, investigators chose a very sensitive definition of bleeding. Had they chosen a TIMI major bleeding definition, there would have been no increased bleeding risk with clopidogrel vs. placebo on a background of ASA, Dr. Mehta noted. The CURE trial also demonstrated that clopidogrel does increase major bleeding, although not life-threatening bleeding. The study also indicated that bleeding risk could be reduced by lowering the dose of concurrent ASA and by withholding clopidogrel for five days prior to CABG.

Turning to the newer antiplatelet trials, Dr. Mehta discussed the TRITON-TIMI 38 study in which the older thienopyridine was compared to a new thienopyridine, prasugrel, in both NSTE-ACS and STEMI patients. Patients were all scheduled for PCI but those with NSTE-ACS were randomized to either antiplatelet arm only after their coronary anatomy was known. (It should be pointed out that a delay in PCI as was imposed on patients with NSTE-ACS in the TRITON-TIMI 38 study deprives patients of the benefits of pre-treatment with clopidogrel as was demonstrated in CURE.) Hard clinical end points established for TRITON-TIMI 38 included CV death, MI and stroke. Patients who received more intensive antiplatelet therapy with prasugrel had about a 20% relative risk reduction in the primary outcome compared to those receiving standard doses of clopidogrel. Most of the benefit of prasugrel occurred early, within the first week of treatment.

The same was true for stent thrombosis, with most of the benefit from prasugrel occurring within the first seven days after randomization. There was an increase in TIMI major bleeding, including life-threatening and fatal bleeds with the novel agent. Fatal bleeds were rare but they were significantly higher in this group, Dr. Mehta observed. Other bleeding outcomes in this trial deserve consideration as well, as rates of TIMI minor bleeds were higher among prasugrel recipients (5%) vs. 3.8% for clopidogrel recipients. As Dr. Mehta noted, one of the main reasons patients stop taking clopidogrel is because of nuisance bleeding—for example, every time they brush their teeth—so minor bleeding is not unimportant when it comes to long-term oral antiplatelet therapy. Bleeding requiring transfusion was also higher in the prasugrel arm of this study and when this cohort underwent CABG, important rates of TIMI major bleeding were observed. In light of this finding, Dr. Mehta cautioned that patients on prasugrel should wait at least seven and possibly 10 days before they undergo a bypass procedure.

I also presented a new analysis of the CURE study in which we analyzed outcomes in TRITON subgroups (prior stroke or TIA, age <u>></u>75 years, weight <60 kg) who did not derive a net benefit from prasugrel, and reported them as net clinical benefit or net clinical harm. Our analysis showed the following:

Table 1.

706" />705" />

Table 2.

<img2706|center>

Table 3.

<img2707|center>

This new analysis of CURE data confirms that there is a consistent net clinical benefit from dual antiplatelet therapy for high-risk as well as low-risk NSTE-ACS patients. Conversely, replacing clopidogrel with prasugrel in ACS patients with a prior stroke or TIA is associated with net clinical harm and does not appear to be any more effective than placebo. Based on analyses of different databases, up to 40% of ACS patients in clinical practice have at least one of these three high-risk features and therefore would not be appropriate candidates for prasugrel therapy.

Summary

Over the next few years, we will welcome a number of new antiplatelet and antithrombotic options for the management of ACS patients, and weighing their efficacy against bleeding risk will be mandatory before adopting their use in routine clinical practice. Not all patients respond equally well to antiplatelet therapy, thus we will need to be guided by the general principle of trying to identify the right strategy for the right patient given at the right time and at the right dose. It is a challenge, but one we must meet to achieve optimum benefit and minimum risk.