Robert A. Byrne, MB, BCh, PhD; Adeel Shahzad, MBBS, MRCP; Rodney H. Stables, DM, FRCP


April 03, 2014

This feature requires the newest version of Flash. You can download it here.

Editor's Note:
This discussion was recorded before the official presentation of the HEAT-PPCI findings at the American College of Cardiology (ACC) Scientific Sessions 2014 in Washington, DC. Both the trial findings and methodology sparked controversy, as covered by heartwire.

HEAT-PPCI: Bivalirudin vs Heparin With Bailout GPI

Robert A. Byrne, MB, BCh, PhD: Hi. I am Robert Byrne from the Deutsches Herzzentrum in Munich. We are here at ACC 2014 in Washington, DC. We just heard some very interesting trial data. Today we are going to talk about 2 studies: HEAT-PPCI[1] and BRAVE-4.[2] BRAVE-4 is a study in which we had some involvement at Munich, and HEAT-PPCI was conducted by the Liverpool group. I am delighted to be joined by Adeel Shahzad and Rod Stables from Liverpool.

First, let's say a few words about HEAT-PPCI.[1] Adeel, can you give us a little background on the design of the study?

Adeel Shahzad, MBBS, MRCP: Thank you very much, Robert. In the primary percutaneous coronary intervention (PCI) setting, bivalirudin and bailout use of glycoprotein (GP) IIb/IIIa inhibitors is an established treatment option. This has never been tested -- bivalirudin against heparin with the same amount of GP IIb/IIIa inhibitor (GPI) effect. HEAT-PPCI is the first trial ever to compare bivalirudin and heparin in the primary PCI setting with the use of GPI bailout, in both conditions.

Dr. Byrne: The novelty in this setting is that you were restricting GPI to bailout use, an approach that makes sense. We conducted the BRAVE-3[3] trial in Munich with ST-elevation myocardial infarction (STEMI) patients. We hypothesized that if you gave people clopidogrel loading (600 mg), then probably you weren't going to add much with GPI use. In that study, we didn't see any benefit against placebo with GPI use. We tend to be quite restrictive with GPIs, at least at our center, which reflects the use in your trial.

Rodney H. Stables, DM, FRCP: I agree. Selective use is increasingly the international norm and is in fact represented in both the European and American guidelines. We are in a unique situation with GPI agents in that we can gain useful information at the time of angiography about the patients who have the greatest potential for benefit, and we can balance this against the perceived bleeding risk. The thrombus burden, the response to simple steps, and intervention will guide us as to which patients probably need the GPI agents. But, of course, the exact optimum frequency of use is yet to be decided.

Dr. Byrne: Yes. These are STEMI patients, and it was an unselected enrollment, or the inclusion criteria for patients with STEMI were fairly broad, is that right?

Dr. Shahzad: This is a true, all-comers trial, and every patient who activated the primary PCI pathway and was having STEMI in the acute-phase management and was planning to have primary PCI as an index procedure was enrolled, if they were eligible. We have successfully recruited 100% of all eligible patients, creating a real-world population.

Dr. Stables: That is quite a significant achievement when you think about the stresses of primary PCI: patients arriving with minimal notice 24 hours a day, 7 days a week, managed by a wide variety of staff manning an on-call rota. We achieved fantastic cohesion of the study's colleagues so that not a single eligible patient over a period of 22 months was missed.

Ischemic and Bleeding Endpoints: Apples and Oranges

Dr. Byrne: The question, then, is quite straightforward. In this setting, bivalirudin or heparin, what was the main endpoint of the study or what did you decide to look at?

Dr. Stables: We looked at the primary efficacy outcome, which was a major adverse cardiovascular event (MACE) composite of all-cause mortality, cerebrovascular accident, recurrent infarction, and additional unplanned target lesion revascularization (TLR). As we showed in the presentation, TLR made almost no contribution to the composite. Only a single TLR event occurred in the absence of a more significant event. We are really looking at the rates of more substantial events here.

Dr. Byrne: These are hard clinical endpoints. You have decided to go for an ischemic endpoint rather than a composite endpoint, including ischemia and bleeding complications. The primary endpoint included ischemic events.

Dr. Stables: Yes. We had a primary safety efficacy outcome, which looked at major bleeding. That is obviously a subject for debate, but a strong body of opinion believes that there are many methodological problems with combining safety and efficacy outcome measures, not least that it creates a composite that, by design, will contain elements that will move in opposite directions.

Dr. Byrne: Exactly.

Dr. Stables: This makes it very difficult.

Dr. Byrne: We have learned a lot in recent years and we have all become more cognizant of the fact that bleeding probably has an impact on mortality that is equal to ischemic endpoints, but it is because they tend to move in different directions. There is certainly some argument about not combining them into a composite endpoint.

Dr. Shahzad: We stuck to the principle of keeping them separate.

More Stent Thrombosis With Bivalirudin

Dr. Byrne: So you had prespecified primary efficacy and prespecified primary safety. What did you find?

Dr. Stables: In terms of the primary efficacy outcome, a statistically significant difference emerged. Adeel, you probably know the figures better than anyone.

Dr. Shahzad: Patients who received bivalirudin had a significantly higher incidence of MACE composite at 8.7% vs 5.7% in patients who received heparin. This difference was mainly driven by the reinfarction events and acute stent thrombosis events.

Acute stent thrombosis events were significantly (4-fold) higher in the bivalirudin group in comparison with the heparin group. According to the Academic Research Consortium (ARC) criteria, they were mainly definite events in the first 24 hours.

Dr. Byrne: So there is a clear message in terms of stent thrombosis driving the difference in the ischemic endpoint.

Dr. Shahzad: Yes. We saw an absolute risk increase of 3% in the bivalirudin arm and a relative risk increase of 1.52.

Dr. Stables: That is the MACE event relative risk. The actual difference in the stent thrombosis relative risk is about 4-fold, which is remarkably similar to the rates seen in EUROMAX[4] and in HORIZONS-AMI,[5] in which the relative risk was actually 5-fold.

Dr. Byrne: It's quite a stark message. Questions that it brings to mind are: What about GPI use? How much GPI was used in these patients? The second question is, what about the adenosine diphosphate (ADP)-receptor antagonist? Were they contemporary modern treatments?

Dr. Shahzad: The use of GPI was similar in both groups, between 13% and 15%. In terms of dual-antiplatelet loading, it was almost universal. The patients received clopidogrel, prasugrel, or ticagrelor, and 60% of the patients received ticagrelor.

No Difference in Bleeding

Dr. Byrne: So you used modern therapy and most patients were treated with ticagrelor, with a relatively low rate of selective GPI use. The next important question is about bleeding. What did you see in terms of bleeding?

Dr. Stables: In terms of major bleeding, which was predefined as Bleeding Academic Research Consortium (BARC) criteria 3-5, there was no statistically significant difference, with rates of approximately 3% in both groups. We also looked at minor bleeding, defined as BARC 2, and the combination of major or minor bleeding. Again, there were no differences.

Dr. Byrne: That is challenging a paradigm that we have come to accept -- that bivalirudin tends to be associated with lower rates of bleeding. How do you account for this?

Dr. Stables: I would like to challenge that. In trials where bivalirudin monotherapy was compared with heparin monotherapy at reasonable heparin doses, no bleeding advantage ever emerges. If we take ourselves back to the ACUITY trial,[6] admittedly in the non-STEMI arena, there were study arms with universal GP IIb/IIIa and with heparin and universal GP IIb/IIIa, and these groups also had identical bleeding. We know from many trials over more than a decade that differential use of GPI results in differential rates of bleeding. Hence we feel that the leading trials that substantiated bivalirudin use in the primary PCI setting were to some extent flawed, in that it is difficult to make a reliable comparison with the relative efficacy and the safety of the 2 agents with a background of differential GPI use. GPI use drives the bleeding.

Dr. Byrne: That is an interesting study. The other study is BRAVE-4, which we were involved in coordinating from Munich. We looked at a similar question -- bivalirudin against heparin in STEMI patients. The bivalirudin patients were also treated with prasugrel and the heparin patients were also treated with clopidogrel. The similarity in both studies was low GPI use. In our study it was approximately 4.5%, a low rate of GPI use. Of interest, in terms of bleeding, we didn't see any difference between the 2 groups, which is in keeping with what you saw. The primary endpoint in our study was a composite of ischemic and bleeding endpoints. Overall, this was also similar in both groups. So there are some consistencies in the message between the 2 studies and some slightly different findings.

Dr. Stables: I don't find the findings that you have described from the BRAVE-4 and HEAT-PPCI studies in any way surprising. They are in keeping with a body of evidence that has developed in a stepwise manner over a protracted period.

Guidelines and Cost-Effectiveness

Dr. Byrne: Let's finish with the question on everyone's mind, which is about stent thrombosis. We now have HORIZONS-AMI, EUROMAX, and your study showing an excess of stent thrombosis with bivalirudin therapy. The guidelines, on the other hand, recommend bivalirudin as the therapy of choice in patients who are undergoing primary PCI. Is it time for a rethink? Do we need to update the guidelines? What are the implications of your study and BRAVE-4 in this respect?

Dr. Stables: I'm listening to data emerging here at the meeting in Washington, DC, and there seem to be signals that possibly bivalirudin results can be improved if the infusion is continued as a high-dose regimen for 4 hours. I can't advocate that strategy, because even if it was possible with a magic wand to abolish excess bivalirudin-related stent thrombosis in HEAT, you would only gain equivalence. The strategy that you suggest would change the drug cost differential from 500-fold to perhaps 1500-fold with the cost of ampules for prolonged profusion to achieve a comparable result. When cost is such an important issue in modern healthcare, society will have to look at this. The guidelines maybe need some revision.

Dr. Byrne: We are all in agreement on that point. Thanks for coming here today. Congratulations on an excellent study, and one which will have a long-lasting impact.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.