HORIZONS-AMI: Bivalirudin Reduces Cardiac Death in Diabetics

Download this article's Factoid (PDF & PPT for Gold Subscribers)

The direct thrombin inhibitor bivalirudin results in significantly reduced rates of cardiac death compared with heparin and a glycoprotein IIb/IIIa inhibitor (GPI) in diabetics with ST-segment elevation myocardial infarction (STEMI) undergoing percutaneous coronary intervention, according to a substudy of the HORIZONS-AMI trial. But, according to 1-year results appearing in the July 2011 issue of JACC: Cardiovascular Interventions, bivalirudin may not reduce overall mortality or major bleeding in this population.

For the HORIZONS-AMI (Harmonizing Outcomes with RevascularIZatiON and Stents in AMI) trial, researchers led by Gregg W. Stone, MD, of Columbia University Medical Center (New York, NY), randomized 3,602 STEMI patients at 123 centers in 11 countries who were undergoing primary PCI to receive bivalirudin (n = 1,800) or heparin plus a GPI (n = 1,802). In a second randomization, 3,006 of the patients were assigned in a 3:1 ratio to receive a paclitaxel-eluting Taxus stent (n = 2,257; Boston Scientific, Natick, MA) or an otherwise identical BMS (n = 749).

At 3 years, bivalirudin significantly reduced major bleeding (non-CABG), reinfarction, cardiac death, and all-cause death compared with heparin plus a GPI, while Taxus patients experienced a 40% reduction in ischemic TLR compared with BMS patients.

Focus on Diabetics

In the new subanalysis, researchers led by Dr. Stone focused on the 593 patients from HORIZONS-AMI who were diabetic. Of these, 281 received bivalirudin and 312 received heparin plus a GPI.  Baseline and procedural characteristics were similar between the 2 groups, although thienopyridine use prior to admission was higher in diabetics receiving heparin plus a GPI compared with those receiving bivalirudin (6.1% vs. 2.5%; P = 0.03).

At 30 days, the rate of cardiac death was reduced in patients receiving bivalirudin. However, noncardiac deaths were higher in these patients, rendering the endpoint of overall death equivalent in both groups. In addition, net adverse clinical events (major bleeding or MACE), MACE (death, reinfarction, TVR, or stroke), TVR, and major non-CABG bleeding all were equivalent (table 1).

Table 1. HORIZONS-AMI: 30-day Outcomes in Diabetics


(n = 281)

Heparin Plus GPI
(n = 312)

P Value

All-Cause Death




     Cardiac Death








Net Adverse Clinical Events








Major Bleeding (Non-CABG)




By 1 year, cardiac deaths remained reduced in the bivalirudin group, while the difference in noncardiac deaths was no longer significant. MACE, and major bleeding remained similar (table 2).

Table 2. HORIZONS-AMI: 1-year Outcomes in Diabetics


(n = 281)

Heparin Plus GPI
(n = 312)

P Value

All-Cause Death




     Cardiac Death












Major Bleeding (Non-CABG)




Net adverse events also were similar at 1 year, though exact rates were not reported. Stent thrombosis rates were likewise equivalent between the bivalirudin (4.2%) and heparin plus GPI (3.8%; P = 0.85) arms. Breaking down the diabetes patients by insulin-treated and noninsulin-treated status, antithrombotic strategy made no difference in net adverse events, MACE, or stent thrombosis. However, among insulin-treated patients, bivalirudin was associated with a lower incidence of cardiac death at 1 year (1.4% vs. 9.4%; P = 0.04).

The current study “represents the largest analysis of diabetic STEMI patients treated by contemporary primary PCI to date and demonstrates that bivalirudin compared with heparin plus GPI significantly reduces cardiac mortality in the high-risk diabetic cohort as well as patients without diabetes,” Dr. Stone and colleagues conclude.

No Interaction with Diabetes Status Found

Because bleeding and overall mortality were not reduced as they were in the main cohort, the authors performed an interaction analysis, demonstrating that the favorable effects of bivalirudin were independent of diabetic status for the endpoints of major bleeding (P for interaction = 0.26), net cardiac events (P = 0.64), all-cause mortality (P = 0.55), and cardiac mortality (P = 0.19). This showed that “the most appropriate interpretation of these data is that the overall benefits of bivalirudin seen in the main trial seem to apply to patients both with and without diabetes,” the researchers write, although they could not rule out a “modest difference in the relative effect of bivalirudin versus heparin plus GPI in patients with, versus without, diabetes.”

According to Debabrata Mukherjee, MD, of Texas Tech University Health Sciences Center (Lubbock, TX), the interaction analysis was the most meaningful part of the substudy. “That’s very important [because] it means statistically, whether the patient has diabetes or not, they get benefit, and having diabetes shouldn’t factor into the decision to use bivalirudin or not,” he told TCTMD in a telephone interview. “This is very reassuring.”

Therefore, Dr. Mukherjee added, the subanalysis results should be viewed in the same context as the overall trial. “Based on the totality of the data, whether the patient has diabetes or not, bivalirudin appears to be the antithrombotic of choice. It’s pretty robust, strong evidence,” he said, adding that in his own practice, “we use bivalirudin irrespective of diabetes in the majority of patients.”

Change Is Slow, But Steady

Nevertheless, clinicians have not made a wholesale switch to bivalirudin. “Physician practice takes time to change. There’s a lot of bivalirudin use in the younger generation, newer physicians and people coming out of training,” Dr. Mukherjee said. “For people who have trained with unfractionated heparin, it takes time, sometimes years or decades, but bivalirudin use has gone up. It’s maybe 50%. It’s not 100%, but it will take time.”

Regarding the question of why the subanalysis failed to show differences in overall mortality, MACE, or major bleeding, Dr. Mukherjee said the answer may lie in the sample size. “If you did a several-thousand patient diabetic trial, I think you’d see a similar magnitude of bleeding reduction with bivalirudin as in the overall trial,” he added.

Note: Dr. Stone and several coauthors are faculty members of the Cardiovascular Research Foundation, which owns and operates TCTMD.


Witzenbichler B, Mehran R, Guagliumi G, et al. Impact of diabetes mellitus on the safety and effectiveness of bivalirudin in patients with acute myocardial infarction undergoing primary angioplasty. Analysis from the HORIZONS-AMI (Harmonizing Outcomes with RevascularIZatiON and Stents in Acute Myocardial Infarction) trial. J Am Coll Cardiol Intv. 2011;4:760-768.



  • Dr. Stone reports serving as a consultant to Abbott Vascular, AstraZeneca, Boston Scientific, Bristol-Myers Squibb-Sanofi, Eli Lilly, Merck, and The Medicines Company.
  • Dr. Mukherjee reports no relevant conflicts of interest.


Related Stories:

Jason R. Kahn, the former News Editor of TCTMD, worked at CRF for 11 years until his death in 2014…

Read Full Bio