One-year data from the Harmonizing Outcomes with Revascularization and Stents in AMI (HORIZONS AMI) showing that STEMI patients undergoing percutaneous coronary intervention fare best with bivalirudin and Taxus stents are maintained at two-year follow-up, according to late-breaking results at Transcatheter Cardiovascular Therapeutics 2009. Data from two new subanalyses of the HORIZONS-AMI trial show that the mortality-reducing effects of bivalirudin monotherapy are limited to the highest-risk patients and that those with lesions in the left anterior descending coronary artery have less major bleeding and cardiac death when given the antithrombin agent compared with heparin plus a glycoprotein IIb/IIIa inhibitor (GPI).

HORIZONS AMI is a randomized multicenter trial that compared bivalirudin to heparin plus a GP IIb/IIIa inhibitor in 3,602 STEMI patients. Of these patients, 3,006 were eligible for randomization, in a 3:1 ratio, to either Taxus paclitaxel-eluting stents or Express BMS.

TCT Course Director Gregg W. Stone, M.D., of Columbia University Medical Center, New York, N.Y., presented the latest findings from the study's stent and drug cohorts.

At two years, bivalirudin (compared to heparin) significantly reduced non-CABG major bleeding by 36% (p<0.001), reinfarction by 25% (p=0.038), cardiac mortality by 41% (p=0.005), and all-cause mortality 25% (p=0.049). The bivalirudin and heparin plus GP IIb/IIIa arms had comparable rates of stent thrombosis, target-vessel revascularization and stroke.

"We had noticed at one year that there was a slight spreading of the curves for reinfarction favoring bivalirudin and that trend continued over two years. Interestingly, reinfarctions continued to accumulate in the heparin plus GP IIb/IIIa inhibitor arm, while the curve was much flatter in the bivalirudin arm," he said, adding it is possible that reinfarction may be affected by early bleeding, much like long-term mortality.

In the stent analysis, use of Taxus vs. Express BMS significantly reduced ischemic TLR and TVR by 42% and 34% (both p<0.001), respectively, with no evidence of late catch-up. Rates of all-cause mortality, CV mortality, reinfarction and stent thrombosis were comparable between the two groups.

When looking at adverse events between one- and two-year follow-up, the incidence of death was significantly lower with Taxus vs. Express (0.8% vs. 1.8%; p=0.04) as was ischemic TLR (2.6% vs. 4.7%; p=0.006) and TVR (3.4% vs. 5.2%; p=0.03). These observations "should be considered hypothesis-generating given their borderline significance and the possible influence of routine angiographic follow-up," Stone commented.

Also noteworthy, he said, is the finding that the lowest cumulative two-year mortality was observed in patients assigned to bivalirudin plus Taxus (3.8%) and the highest was seen in those randomized to heparin plus a GP IIb/IIIa inhibitor and Express (6.1%).

Commenting on the trial, Christian W. Hamm, M.D., of Kerckhoff Clinic, Bad Nauheim, Germany, said HORIZONS AMI is "indeed a landmark study, and we're very pleased to hear that the two-year follow-up results are very consistent with what we saw at one-year, without any signs of catch-up."

Hamm noted that the lower rate of reinfarction between years one and two in the bivalirudin arm is "surprising."

Stone replied that the reason for the difference is unclear but could be attributed to either bleeding or chance. "You would have to see it replicated in another large trial before we could really believe it," he said.

In the original trial, researchers randomized 3,602 STEMI patients undergoing primary PCI to bivalirudin monotherapy or unfractionated heparin plus a GPI within 12 hours of symptom onset. Bivalirudin was shown to reduce rates of major bleeding and mortality as well as net adverse clinical events compared with heparin plus a GPI.

For one of the new studies, Guido Parodi, M.D., Ph.D., of Careggi Hospital (Florence, Italy), and colleagues stratified patients by risk, classifying them as low (score 0-2), medium (score 3-5), or high (score > 5) risk according to the CADILLAC risk score based on 7 clinical variables. Of 2,530 evaluable patients, 1,522 (60%) were considered low risk, 531 (21%) were medium risk, and 477 (19%) were high risk.

At 1 year, major bleeding was significantly lower with bivalirudin vs. heparin plus a GPI only in the low-risk group (3.0% vs. 6.9%; p=0.0005), whereas bivalirudin lowered mortality only in the high-risk group (8.4% vs. 15.9%; p=0.01). In addition, bivalirudin lowered the incidence of MI in high-risk patients (3.6% vs. 7.9%) while showing a trend toward reductions in several other clinical endpoints in the high-risk group, including stent thrombosis (2.2% vs. 4.0%), MACE (18.2% vs. 25.4%), and net adverse clinical events (27.1% vs. 34.5%) compared with heparin plus a GPI. No differences were seen in rates of TVR and stroke.

"Contrary to our expectations, we found that the benefit of bivalirudin in the treatment of STEMI patients undergoing primary PCI was actually greatest in the highest-risk population of patients, as defined by the well-verified CADILLAC risk score," Dr. Parodi said.

"When randomized to bivalirudin, the relative reduction in mortality for those at highest risk of dying was nearly 50%, with an absolute reduction of 7.5%. [This finding suggests] that for every 100 high-risk patients treated with bivalirudin rather than heparin plus a GP IIb/IIIa antagonist, 7 more patients will be alive at one year," Parodi said.

"Ongoing analyses are designed to determine if this mortality benefit in the patients at highest risk for mortality is due to a greater risk for, and reduction in, bleeding in this cohort," Dr. Parodi said. Meanwhile, "our results can help guide clinicians in their choice of antithrombotic therapies in the treatment of STEMI patients being treated with primary PCI," he added.

In another HORIZONS-AMI subanalysis, Jochen Wohrle, M.D., of the University of Ulm, (Ulm, Germany), and colleagues found that within the subgroup of patients undergoing PCI of the LAD, anticoagulation with bivalirudin (n = 699) resulted in significantly less major bleeding (6.5% vs. 10.1%; p=0.01) and cardiac death (3.0% vs. 5.3%; p=0.04) at 1 year compared with heparin plus a GPI (n=746). Other clinical endpoints including all-cause death, reinfarction, and definite or probable stent thrombosis were equivalent in LAD patients who received bivalirudin or heparin plus a GPI.

Moreover, "there was no difference in clinical outcome between patients with transfer for primary PCI and patients with direct admission at the study hospital with a PCI facility," Wohrle said. "In both groups, the use of bivalirudin was associated with lower rates of bleeding, cardiac death, and net adverse clinical events," he added, noting that the study results strongly support transfer of STEMI patients for primary PCI care.