Vienna, Austria - Some observations of interest from the one-year results of the ACUITY trial were presented in two papers at the European Society of Cardiology (ESC) Congress 2007. These included a trend toward a lower mortality in the bivalirudin group by 12 months and new data suggesting that major bleeding is related to the death rate at one year. Other data shed more light on the clopidogrel debate and whether it is safe to switch from heparin to bivalirudin.

Dr Harvey White (Green Lane Hospital, Auckland, New Zealand) presented new data from the ACUITY patients who had undergone PCI at one "clinical-trial update" session, while Dr Walter Desmet (University Hospital Gasthuisberg, Leuven, Belgium) presented data from the patients who were treated medically at a separate session. But discussants of both presentations commented that the ACUITY trial had several limitations and that these subgroup analyses should be interpreted with caution.

The ACUITY trial evaluated the optimum upstream treatment of patients with moderate- to high-risk non-ST-elevation ACS heading to the cath lab. In the study, 13 819 such patients were randomized to one of three arms: unfractionated heparin (UFH) or enoxaparin plus routine GP IIb/IIIa inhibition; bivalirudin plus routine GP IIb/IIIa inhibition; or bivalirudin alone. Results showed that the bivalirudin-monotherapy group did best, with significantly less bleeding and without a significant increase in events compared with the heparin-plus-IIb/IIIa-blocker arm.

PCI cohort

Presenting data from the subgroup of 7789 patients who ended up undergoing PCI in this study, White focused on three issues in this cohort: patients who were switched from heparin/enoxaparin to bivalirudin at the time of randomization; whether clopidogrel pretreatment is necessary to see the benefit with bivalirudin; and whether bleeding rates correlate with long-term mortality.

Safe to switch

White noted that switching antithrombin agents has fallen out of favor since the SYNERGY trial showed a worse outcome in patients who were crossed over after randomization from heparin to enoxaparin or vice versa. But he presented new data from the ACUITY PCI subgroup suggesting that it was safe to switch from heparin/enoxaparin to bivalirudin. In the 2528 patients in the PCI cohort who were already receiving heparin or enoxaparin at the time of randomization and were then randomized to continue on that treatment or switch to bivalirudin, White reported that the outcome results were similar to the main results of the trial, showing a similar rate of ischemic events but a reduction in bleeding with bivalirudin.

ACUITY: HR (bivalirudin vs heparin/enoxaparin plus IIb/IIIa blocker) of ischemic events and major bleeding in PCI patients who received heparin/enoxaparin before randomization

End point	HR (95% CI)
Composite ischemic end point	1.10 (0.85-1.42)
Major bleeding	0.52 (0.36-0.74)

Clopidogrel pretreatment

White noted that that one controversy about the ACUITY trial has centered on a possible interaction with clopidogrel pretreatment, suggesting that the benefit of bivalirudin may not have been seen in patients not pretreated with the antiplatelet drug. But he pointed out that this interaction was not statistically significant (p=0.054), and he presented new data from the PCI subgroup that he said showed no difference in one-year mortality between the 3429 bivalirudin patients who had been pretreated with clopidogrel and the 1044 who had not been pretreated. "Too much has been made of this possible interaction, and it has not been backed up by the one-year mortality results," he said.

ACUITY: Effect of clopidogrel pretreatment on hazard ratio for one-year mortality (bivalirudin vs heparin/enoxaparin plus IIb/IIIa blocker)

Group	HR (95% CI)
Clopidogrel pretreatment (n=3429)	1.02 (0.69-1.50)
Clopidogrel after end of angiography to 30 min after PCI (n=1044)	1.14 (0.89-2.03)
Clopidogrel after 30 min post-PCI (n=519)	0.43 (0.17-1.11)
No clopidogrel (n=88)	3.20 (0.34-31.1)

Medically treated patients

Focusing on the subgroup of patients who did not undergo any invasive treatment after angiography (about one third of the total ACUITY population), Desmet noted that these patients were a heterogeneous group, tending to be younger and more likely to be female than the invasively treated patients. The medically treated patients were also less likely to have raised biochemical markers or ST changes but had more previous cardiac events at baseline. In the trial follow-up, the medically treated patients had fewer ischemic events and less major bleeding than invasively treated patients. Desmet noted that results in this medically treated cohort showed similar results to the overall trial, with a reduction in major bleeding in the bivalirudin-treated patients at 30 days and similar ischemic and mortality outcomes at one year as compared with patients receiving heparin/enoxaparin plus a IIb/IIIa blocker.

ACUITY: One-year results in medically managed patients

End point	UFH/enoxaparin+IIb/IIIa blocker (n=1493) (%)	Bivalirudin (n=1502) (%)	p
30-d major bleed	4.4	2.5	0.004
1-y composite ischemic end point	9.0	8.7	NS
1-y mortality	4.0	3.8	NS

Bleeding tracks mortality

Both White and Desmet presented data suggesting that bleeding at 30 days was related to mortality at one year. In both the PCI patients and the medically treated subgroups, patients who had a major bleed at 30 days had a greater hazard ratio for one-year mortality than patients who had had an MI at 30 days. "These data show the prognostic importance of major bleeding in ACS patients," Desmet said. White also pointed out that "patients who had PCI and a major bleed spent two extra days in the hospital compared with those who did not bleed, which is a very important direct patient benefit."

ACUITY: Influence of major bleeding and MI in the first 30 days on the risk of death at one year: Medically treated patients

Outcome	HR for death at 1 yr	p
Major bleed at 30 days	3.77	<0.0001
MI at 30 days	3.39	0.0012

ACUITY: Influence of major bleeding and MI in the first 30 days on the risk of death at one year: PCI patients

Outcome	HR for death at 1 yr	p
Major bleed at 30 days	3.16	<0.0001
MI at 30 days	2.30	<0.0001

One-year mortality

Both speakers also pointed out that while mortality results at 30 days showed a small trend in favor of the heparin/enoxaparin-plus-IIb/IIIa-blocker group, this had reversed by one year, with bivalirudin patients now showing a small trend toward a better mortality outcome. This was seen in the overall ACUITY population and in both the PCI and medically treated subgroups. White commented to heartwire: "While these findings are not statistically different, the reversal of the point estimates by one year is interesting and could be due to the reduction of major bleeding with bivalirudin."

ACUITY: One-year mortality results

1-y mortality	Heparin/enoxaparin+IIb/IIIa blocker (%)	Bivalirudin (%)	p
Overall ACUITY population	4.4	3.8	NS
Medically treated patients	4.0	3.8	NS
PCI patients	4.0	3.2	NS

Limitations of ACUITY

Discussant of Desmet's presentation, Dr Robbert de Winter (Academic Medical Center in Amsterdam), said the ACUITY trial had several limitations, including the fact it was a very complex trial, it was open label in design, its noninferiority boundary was too liberal, and the use of GP IIb/IIIa blockers was too short; these limitations must be added to the normal caveats taken into account when subgroup analyses such as these are interpreted.

Which antithrombin to choose?

De Winter noted that the new ESC and the ACC/AHA guidelines on NSTE-ACS, just published, recommend bivalirudin when an early invasive approach is chosen but say it is not tested in patients managed with a more conservative approach, in whom fondaparinux is preferred, adding that the writing committee of the ACC/AHA NSTE-ACS guidelines state that with all the trial information available, it is still very difficult to clearly recommend one treatment regimen over the other.

To heartwire, De Winter commented: "A clear take-home message could be that hospitals and physicians are advised to make decisions on what regimen they want to adopt. They should not use different protocols with different medical regimens simultaneously, as this could result in mistakes in doses and combinations of drugs with the accompanying risk of bleeding."

Responding to this, Desmet pointed out that the decision as to which antithrombin to use was normally made before it was known whether the patients would be treated invasively or not (which was often not decided until angiography). "We have shown that a treatment strategy with bivalirudin is safe and effective also in patients who are eventually triaged to medical management. And if data in 13 819 patients show no significant mortality differences, with a numerical difference in mortality in favor of bivalirudin at one year, together with a significantly decreased incidence in major bleeding (whatever bleeding scale is used), why should that not be enough to recommend bivalirudin for this patient group?"