New Bivalirudin/Heparin Meta-Analysis Offers Consistency of Signals

Shelley Wood

August 14, 2014

BOSTON, MA ( updated ) — Authors of a new meta-analysis comparing a bivalirudin (Angiomax, the Medicines Company)-based regimen vs a heparin-based approach in patients undergoing PCI say they hope their paper helps the medical community to understand the "consistency of the signals" across the major trials[1]. The meta-analysis, published today in the Lancet, includes this year's bivalirudin barn burner, HEAT-PPCI , as well as a number of smaller trials that have been presented at cardiology meetings but are not yet in print, including BRIGHT and NAPLES III .

As previously reported by heartwire , HEAT-PPCI, unveiled at the ACC 2014 meeting, was an open-label, single-center STEMI trial that randomized patients to bivalirudin or low-dose unfractionated heparin, with GP IIb/IIIa inhibitors used in both arms for bailout only. The trial, which has since been published, showed a significantly lower rate of major adverse cardiac events (MACE) in the heparin-treated patients at 28 days and no differences in bleeding complications.

Speaking with heartwire , Dr Marc Sabatine (Brigham and Women's Hospital, Boston, MA), who coauthored the new meta-analysis with Dr Matthew Cavender (Brigham and Women's Hospital), acknowledged the debate that swept into the wake of HEAT-PPCI's presentation and said he hopes this meta-analysis will help people understand why the trial was not such an outlier after all—a point that's been repeatedly made by HEAT-PPCI's senior investigator, Dr Rod Stables.

Lancet Meta-Analysis Results

The meta-analysis included a total of 34 000 patients in 16 trials that included by a mix of elective and/or urgent PCI, NSTE-ACS, STEMI, or AMI populations (as per study definitions). In all, 2422 experienced a major adverse cardiac event (MACE) within the first 30 days, and 1406 had a major bleed, according to different bleeding definitions used across the different trials.

When all of the studies were considered together, a bivalirudin-based regimen was associated with a statistically significant, increased risk of MACE, MI, ischemia-driven revascularization, and stent thrombosis, but also with a reduced risk of bleeding. Stent-thrombosis rates were particularly pronounced for acute stent thrombosis and particularly in patients with STEMI.

All Trials: Main Results

End point Hazard ratio 95% CI p
MACE 1.09 1.01–1.17 0.0204
MI 1.12 1.03–1.23  NA
Ischemia-driven revascularization 1.16 0.997–1.34  
Mortality 0.99 0.82–1.18  
Stent thrombosis 1.38 1.09–1.74 0.0074
Major bleeding 0.62 0.49–0.78 <0.0001
NA=not available

When trials were stratified based on whether GP IIb/IIIa inhibitors were used predominantly in the heparin arm only, provisionally in both arms, or planned in both arms, the bleeding benefit was seen only for the first, but not the second or third, comparisons.

Risk for Major Bleed by GP IIb/IIIa Inhibitor Usage

GP IIb/IIIa inhibitor use Hazard ratio for major bleeding 95% CI p
Predominantly in heparin arm 0.53 0.47–0.61 <0.0001
Provisional in both arms 0.78 0.51–1.19 0.25
Planned in both arms 1.07 0.87–1.31 0.53

"I think the increased risk of MACE has been there for virtually all the trials, [although] not all of them statistically significant but certainly numerically, showing an excess in the bivalirudin arm, and so in that sense HEAT-PPCI is not an outlier for that," Sabatine commented. "The results for stent thrombosis, and in particular for acute stent thrombosis in STEMI, wind up being quite consistent with a risk that's roughly fourfold in HORIZONS , EUROMAX , and HEAT-PPCI. So seeing the consistency of the data beyond one very well-done trial and ranging from small to large trials, all showing a consistent message, I think that is useful."

Likewise, he continued, the meta-analysis "also helps put in context the bleeding issue. I think people have always struggled with how to interpret the bleeding results, not only based on anticoagulation but also GP IIb/IIIa inhibitor use. By stratifying the trials into groups [according to GP IIb/IIIa inhibitor use], this provides a clearer message and one that fits with our sense of biology: that when you have differential use of antithrombotics, you have differences in bleeding, and that when you get to more similar [antithrombotic] regimens, the bleeding differences will go away."

Another Meta-analysis

The Lancet paper is actually the second meta-analysis tackling this topic to emerge this month. On August 8, EuroIntervention published a meta-analysis by Cassese and colleagues that included 10 trials but omitted those in which GP IIb/IIIa inhibitor use was planned for the heparin arm[2]. In this analysis (unlike the Lancet's), the reduction in bleeding with bivalirudin was statistically significant, although the authors acknowledge a high degree of heterogeneity across studies. Bivalirudin was, again, associated with a roughly twofold increased risk of stent thrombosis, a finding that was more homogenous across the studies.

Commenting on both studies to heartwire by email, Dr Adnan Kastrati, senior author on the EuroIntervention paper, observed: "Both the Lancet and our EuroIntervention meta-analyses are saying the same thing at the end," the key difference being Kastrati and colleagues' decision to include only trials with provisional GP IIb/IIIa inhibitor use, since this better reflects modern-day practice, he said.

A similar approach was taken in a post hoc analysis of EUROMAX, looking only at patients who received provisional GP IIb/IIIa inhibitors.

The problem with these after-the-fact analyses, said Sabatine, is that the results are confounded by the fact that physicians were already taking into account the patients' likelihood of bleeding or having an ischemic event when they choose to add a GPIIb/IIIa inhibitor.

A second issue is dosing, he continued, both the heparin dose, which ranged from <70 U/kg to 140 U/kg across the trials in the Lancet meta-analysis, and the bivalirudin dose—or rather, how long its infusion was continued. One of the theories put forward in recent months is that prolonging bivalirudin infusion could help cut the ischemic risk. According to Sabatine, this will be an "interesting theory to explore" but is as yet unproven.

Stables, for his part, has pointed out that this strategy would be radically more expensive than a heparin-based regimen, but Sabatine believes that doesn't mean it shouldn't be investigated.

"As a scientist, I focus first on [achieving] the right balance of thrombotic and bleeding events, and when that's been defined, we can figure out what's the most cost-effective, because costs will always change over time."

Meanwhile, in the Cath Lab

Asked what impact these meta-analyses and HEAT-PPCI will have on cath labs—some of which are already making a switch back to heparin—both Kastrati and Sabatine hedged their bets.

"The increased risk of acute stent thrombosis is in fact shown only for STEMI," Kastrati pointed out. "Although these results might have a negative impact even on the use of bivalirudin in non-STEMI, there is still undisputed favorable evidence for bivalirudin in non-STEMI patients." He also says he has been "very happy" to have had bivalirudin for both routine use in the cath lab and to be able to study it across different clinical scenarios.

Nevertheless, Kastrati predicts, "the recent data and the higher price of bivalirudin will probably lead to drastic reductions in the use of this drug in STEMI patients."

Sabatine, likewise, couldn't say whether use of bivalirudin use had changed at Brigham. "If the question is, which drug would I choose in which patient, in those patients in whom you are more concerned about ischemic or thrombotic events you may be more likely to choose the regimen that might prevent those." The risk factors for bleeding are also already "appreciated," namely advanced age, low body weight, and chronic kidney disease, "and in those patients, if they are low risk for ischemic and thrombotic events, then you might want to choose a safer regimen."

Dr Michael Lincoff (Cleveland Clinic, OH), who wrote the accompanying Comment in the Lancet, took a somewhat firmer stand[3].

"Sometimes old and inexpensive drugs are nevertheless good drugs, and sometimes improvements in medical practice can reinvigorate old drugs," he wrote. "Despite its mechanistic disadvantages, this might be the case with heparin. Unless new trial data to the contrary with bivalirudin become available, the findings of this meta-analysis provide a rationale to consider heparin monotherapy as the preferred anticoagulant regimen for contemporary PCI."

The next trial to offer new insights will likely be MATRIX , Lincoff noted.

The ambitious Italian 6800-patient trial is looking at transradial vs transfemoral PCI, bivalirudin vs heparin (with provisional GP IIb/IIIa inhibitors), and periprocedural bivalirudin vs a prolonged bivalirudin infusion. MATRIX results aren't expected until the end of 2014, Lincoff predicts.

As for Stables, he told heartwire that he needed more time to study the meta-analysis in detail but that his first impressions "are favorable" and that they "mirror [his] general reading of the evidence base."

"Obviously. I am pleased that the main findings are concordant with the results of our recent HEAT-PPCI trial. I have maintained for many years that bivalirudin is . . . inconvenient to use, very expensive, and [a] marginally less effective alternative to heparin. It is interesting to reflect how bivalirudin emerged as a 'big seller' so late in the natural history of its market availability. I am sure that this exercise would serve to illustrate that it may be unwise to 'tinker' with some of the core concepts of the scientific method and central tenets of clinical trial methodology."

Sabatine disclosed that he has received research grant support through Brigham and Women’s Hospital from Abbott Laboratories, Accumetrics, Amgen, AstraZeneca, Bristol-Myers Squibb, Critical Diagnostics, Daiichi-Sankyo, Eisai, Genzyme, GlaxoSmithKline, Intarcia, Merck, Nanosphere, Roche Diagnostics, Sanofi, and Takeda; he has done consulting for Aegerion, Amgen, AstraZeneca, Bristol-Myers Squibb, Cubist, GlaxoSmithKline, Intarcia, Merck, MyoKardia, Pfizer, Quest Diagnostics, Sanofi, Vertex, and Zeus Scientific. Cavender declared no competing interests. Kastrati has disclosed submitting patents in relation to a number of drug-eluting-stent technologies and has received consulting or lecture fees from Abbott, Biosensors, and Biotronik. Lincoff disclosed receiving grants from Lilly, AstraZeneca, CSL, Roche, Genentech, Regado, Pfizer, Takeda, and Vivus.


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.
Post as: