This article is more than 5 years old. Information may no longer be current.
HEAT-PPCI: Heparin superior to bivalirudin in 28-day outcomes after primary PCI
WASHINGTON — Results from a single-center study of patients with STEMI undergoing primary PCI showed that heparin was associated with reduced rates of MACE and stent thrombosis compared with bivalirudin at 28 days.
Rod Stables, MD, and colleagues randomly assigned 1,830 consecutive patients presenting with STEMI in 2012 and 2013 at Liverpool Heart and Chest Hospital, Liverpool, United Kingdom, to assess whether use of unfractionated heparin or bivalirudin led to better outcomes when glycoprotein IIb/IIIa inhibitors were not planned to be used in a real-world setting.
Clinically relevant effect
“We concluded that the application of heparin rather than bivalirudin, both used in the context of guideline-endorsed, highly selective (glycoprotein IIb/IIIa inhibitor) use, reduced the rate of major adverse events, however you wish to describe them,” Stables said. “This was a significant effect not only in terms of statistical testing, but in terms of clinical relevance, with as few as 33 people needed to be treated to prevent one event.”
Previous studies comparing heparin with bivalirudin in patients undergoing PCI for STEMI were confounded by differential use of glycoprotein IIb/IIIa inhibitors, so in the HEAT-PPCI study, glycoprotein IIb/IIIa inhibitors were reserved only for “bail out” use in both groups, Stables said at a press conference during the American College of Cardiology Scientific Sessions.
Patients received dual antiplatelet therapy before PCI and were randomly assigned to 70 units/kg body weight of heparin pre-procedure or a bolus of bivalirudin 0.75 mg/kg followed by infusion of bivalirudin 1.75 mg/kg/hour for the duration of the procedure. Follow-up was 28 days.
The primary efficacy outcome was MACE, defined as all-cause mortality, cerebrovascular accident, reinfarction or additional unplanned target lesion revascularization. The primary safety outcome was rate of major bleeding, defined as type 3 to 5 in the Bleeding Academic Research Consortium definitions.
MACE occurred in 8.7% of patients assigned bivalirudin compared with 5.7% of patients assigned heparin (absolute risk increase, 3%; 95% CI, 0.6-5.4; RR=1.52; 95% CI, 1.1-2.1). The results were driven primarily by reinfarction and TLR, Stables said.
Those assigned heparin also had lower rates of stent thrombosis compared with those assigned bivalirudin (0.9% vs. 3.4%; RR=3.91; 95% CI, 1.6-9.5).
There was no difference between the groups in bleeding complication rates (heparin group, 3.1%; bivalirudin group, 3.5%; RR=1.15; 95% CI, 0.7-1.9).
Controversy over study design
Stables noted that the study generated some controversy because of its delayed consent design, in which patients were not asked for informed consent to enroll in the study until after the procedure was performed. Although some have found this unethical, the patients did not, because only three of them refused consent, and all others supplied 28-day data, he said.
“Later, when they were well, they were approached for full informed consent in the traditional manner,” Stables said. “The greatest endorsement of the legitimacy and reasonableness of our approach is that we spoke to nearly 1,800 patients in this setting and only three expressed disapproval and would not consent. In fact, the refusal to consent was not due to disapproval of the trial, but relating more to social issues.” – by Erik Swain
For more information:
Stables R. Late-Breaking Clinical Trials V: TCT@ACC-i2. Presented at: American College of Cardiology Scientific Sessions; March 29-31, 2014; Washington, D.C.
Disclosure: Stables reports financial disclosures with AstraZeneca and The Medicines Company.
Perspective
Back to Top
HEAT-PPCI is an interesting study in that it was an all-comers trial addressing an important issue: the most appropriate anticoagulation in STEMI. A limitation is that it was a single-center study, so it’s applicable to a single center and the exact way they did all the study procedures.
The results are in contradistinction to what we’ve seen with other randomized trials of bivalirudin compared with heparin either with or without glycoprotein IIb/IIIa inhibitors in multicenter trials. Specifically, there have been three other such multicenter trials, which have shown for the most part similar or improved ischemic event rates with bivalirudin, a reduced rate of bleeding and a higher rate of acute stent thrombosis within the first 24 hours, with no difference in acute stent thrombosis after the first 24 hours.
What we saw in HEAT-PPCI was quite different. We saw no difference in bleeding between the two arms and a marked increase in stent thrombosis with bivalirudin — a three to four times higher rate than we saw in either of the large multicenter HORIZONS-AMI or EUROMAX trials. In addition, the required rate of “bail out” glycoprotein IIb/IIIa inhibitors was also significantly higher. This center is known for its very rapid door-to-balloon times. The investigators measured activated clotting time (ACT); we haven’t seen those data yet, but they mention that they are substantially lower than what we’ve seen in any other study of bivalirudin, and very similar to the heparin arm. What we’ve seen in prior studies is that bivalirudin usually has a 100- or 200-second higher ACT than heparin; despite this, it has less bleeding. Here, the ACT was similar in both arms. They did use a different type of assay, which is not as well-validated, and it has a lower ACT level; but nonetheless, the ACTs were similar between bivalirudin and heparin.
This suggests to me that perhaps at the time they did the PCI, they had double the rate of glycoprotein IIb/IIIa inhibitor “bail out” we’d expect because the ACTs were lower; they also had a three-times-higher rate of acute stent thrombosis, possibly because they may have inadvertently used too little bivalirudin. This is why you’ve got to be careful about generalizing the results of single-center trials to worldwide or multicenter experiences. We need more data and to see this manuscript in print after it undergoes peer review to fully understand their processes.
The benefits of bivalirudin in reducing cardiac mortality in HORIZONS-AMI continued to spread over time. There was even a reduction in reinfarction with bivalirudin. So we certainly need longer-term data as well.
There have been multiple meta-analyses of primary angioplasty with unfractionated heparin as the anticoagulant with or without glycoprotein IIb/IIIa inhibitors. Those studies were pretty consistent in that there was a reduction in mortality and infarction with glycoprotein IIb/IIIa inhibitors. In the CADILLAC trial, there was also a reduction in stent thrombosis, even though there was increased bleeding. For that reason, it seems like a pretty good trade-off. At least in the United States, almost no one would use heparin alone without a glycoprotein IIb/IIIa inhibitor for primary PCI. In Europe, primarily because of cost reasons, that’s not necessarily the case, and heparin alone is still a widely used regimen. I believe that’s why they made the decision to test bivalirudin alone vs. heparin alone. It’s not that it’s an uninteresting or invalid scientific question; I just have concerns that they did not use bivalirudin with either sufficient dose or duration to have reduced acute ischemic complications.
Why the bleeding was not less is very difficult to understand. We need to understand more about how they ascertained and adjudicated the complications. It’s always difficult [to draw conclusions from] a single-center trial, which is why we usually rely on multicenter trials. A similar example might be the TAPAS trial, which was a large single-center trial of thrombus aspiration in primary angioplasty suggesting a mortality reduction. When a much larger multicenter study — the TASTE trial — was done, there was absolutely no difference in mortality.
Perspective
Back to Top