The Short-Term Aggrastat Infusion Registry (STAIR): Evaluating the Safety and Efficacy of Heparin + Short-Term Tirofiban vs Bivalirudin in PCI

Operators now have several anticoagulation options available for use in percutaneous coronary intervention. How does the STAIR registry¹ help to provide some clarity?

When we first started doing percutaneous coronary intervention (PCI) when I was a brand-new fellow, aspirin and heparin were our treatment strategies, but we still noted that a high percentage of patients ended up having clots form and all sorts of other problems. We needed something more. Basic scientists started studying platelets, a major cause of thrombosis in the coronary arteries, and identified a glycoprotein (GP) IIb/IIIa receptor that turned out to be the final common activation pathway in platelet aggregation. The first GP IIb/IIIa inhibitor was abciximab (ReoPro), which binds permanently to the GP IIb/IIIa receptor. We added abciximab to acute coronary syndrome patients and even stable patients, with a significant reduction in ischemic complications after PCI. However, there was a significant increase in bleeding complications. Several studies went on to show that if a GP IIb/IIIa receptor inhibitor is on board, we don’t need to give as much heparin. We started using less heparin and experienced less bleeding, but there were problems with abciximab. Abciximab induces antibodies that can react with platelets — the platelet count can plummet and cause thrombocytopenia, and if your patient has bleeding complications, abciximab takes more than a day to wear off. People worked to find other GP IIb/IIIa receptor antagonists, and eptifibatide (Integrilin) and tirofiban (Aggrastat) were tested, found to be beneficial, and approved by the FDA. Both eptifibatide and tirofiban were also less expensive than abciximab. At our center, we generally used eptifibatide until the P2Y12 receptor antagonists came along: clopidogrel (Plavix), which was followed by prasugrel (Effient) and ticagrelor (Brilinta). P2Y12 receptor antagonists are very effective in reducing stent thrombosis and worked during the acute phase as well. The question became, is a GP IIb/IIIa receptor antagonist needed if we are already using dual antiplatelet therapy with a P2Y12 receptor inhibitor? We also had the option of bivalirudin, which is a direct thrombin inhibitor and doesn’t activate platelets like heparin does. At Intermountain, we had significant experience with GP IIb/IIIa receptor antagonists and didn’t particularly want to stop their use. We also did not want to preload all patients with dual antiplatelet therapy, because if patients end up going to surgery, they might have significant bleeding complications. I never did fully jump on the bandwagon for the use of bivalirudin, but some people in our institution did. It turned out that almost half of our physicians used bivalirudin and the other half used eptifibatide. When eptifibatide pricing went up, Medicure offered tirofiban at a more cost-effective price than eptifibatide. Our pharmacy wanted to lower costs and contacted us in cardiology to see if we could determine any differences between eptifibatide and tirofiban. We ultimately concluded that tirofiban and eptifibatide work in an almost identical manner, so we agreed to make the switch to tirofiban.² Then came the idea that maybe once dual antiplatelet therapy is on board, the recommended long-term infusions of a GP IIb/IIIa receptor antagonist aren’t necessary. It was self-evident that continuing the long-term infusion meant patients were more likely to have increased bleeding complications. So, perhaps if patients were loaded on dual antiplatelet therapy with aspirin and a P2Y12 inhibitor, we might not need to leave them on a long-term infusion of a GP IIb/IIIa receptor antagonist. The BRIEF-PCI trial looked at reducing the long-term infusion, concluding that at least in stable patients, it could be done.³ We embraced that result and were impressed to find a low incidence of bleeding complications when using a short-term eptifibatide infusion or just bolus alone if patients were loaded with dual antiplatelet therapy. We went on to publish a paper concluding this strategy worked well and led to significantly fewer bleeding complications.⁴ At our center, we have been using tirofiban as a short-term infusion unless the patient is very high risk, such as when there is a high intraluminal thrombus burden, and then we use the long-term infusion.⁵ There were also patients at our center who had been receiving bivalirudin instead, contemporaneous with tirofiban patients. We decided to do an observational analysis. We went on to compare patients with coronary artery disease who underwent PCI having received either scheduled, planned tirofiban with short-term infusion or planned bivalirudin, which most of the time was without any use of a GP IIb/IIIa receptor antagonist. We analyzed the data to determine safety and efficacy outcomes (Table 1). STAIR showed very low bleeding complications in both groups, significantly lower than in the past, but there were even lower bleeding complications in the heparin (ACT targeted of ~200 seconds) plus tirofiban group. This group received either bolus alone or bolus plus short-term infusion, meaning 4-6 hours or less of tirofiban before it was stopped, resulting in lower bleeding than the bivalirudin group. Overall major adverse cardiac events (MACE) were fairly similar between bivalirudin and short-term infusion of tirofiban, but the tirofiban group demonstrated a significant reduction in mortality. The HEAT-PPCI study⁶, published in 2014 (STAIR patient data spanned the period of January 1, 2013 to October 11, 2016), concluded that perhaps bivalirudin was not much different than heparin, and so bivalirudin, because it was expensive, lost its luster. Today, we don’t use much bivalirudin anymore, but we do use quite a bit of tirofiban. There was about a 13% incidence of a GP IIb/IIIa antagonist use in some of the bivalirudin studies^6-9 — in which the plan was to not use these antagonists, but when patients ran into trouble, operators ended up using it anyway. I don’t want to have to bail out patients. I want them to do well right from the beginning. If I give a bolus alone or a bolus with a very short infusion, the risk of bleeding complications is extremely low and it eliminates most risk of having to do a bailout, because patients are already safely on a GP IIb/IIIa receptor inhibitor.

What was the categorization of bleeding complications used in the STAIR registry?

We used TIMI major bleeding, which mostly includes signs of overt bleeding, the need for a procedure to stop the bleeding, or hemoglobin drop of a certain amount. We also looked at TIMI minor bleeding as well, so STAIR was based on both TIMI major and minor bleeding.

Should radial access and its reduced bleeding complications offer more confidence to the interventionalist when considering use of a GP IIb/IIIa receptor inhibitor?

There was no TIMI major bleeding in any STAIR registry patient who underwent a radial procedure, whether they received bivalirudin or tirofiban. In that respect, it should give people more confidence that they can use a GP IIb/IIIa receptor inhibitor when performing radial access PCI. I would emphasize that a recent, randomized trial from Canada showed that there was no reduction in mortality in patients who underwent radial access for ST-elevation myocardial infarction (STEMI).¹⁰ You can still do femoral when it is appropriate, but patients generally like radial more than femoral, and when it is easy and appropriate to use radial access, we should do so. It certainly ought to give you more confidence in the use of tirofiban because of the reduced bleeding complications of the radial procedure.

How should the STAIR registry be incorporated into daily practice?

People should consider GP IIb/IIIa receptor antagonists again. We have now found a way that they are safe and effective to use, and tirofiban is quite inexpensive. Giving the drug up front might help avoid a bailout in 12-15% of cases because the patient gets into trouble.⁶ Give a bolus, and if all goes well, then stop it. You are not going to have increased bleeding complications, at least compared to bivalirudin. Some people preload with dual antiplatelet therapy and justify it by saying that hardly any patients go to surgery anymore, and therefore, it is not a big problem. I wouldn’t say that there is only one right way to do things, but I would like everyone to read the STAIR registry publication, think more positively about GP IIb/IIIa receptor antagonists, the drugs that in some respect saved PCI when they came out early in the 1990s, and consider that there is still an advantage to their use. I give tirofiban to every single one of my PCI patients, mostly as a bolus. It works well.

Can you share more about why you give a GP IIb/IIIa receptor inhibitor to all your patients?

I am convinced that it is safe and I expect that a single bolus of a GP IIb/IIIa receptor antagonist in the cath lab probably doesn’t increase bleeding complications, because it wears off very rapidly after the procedure is over, anywhere from 90 minutes to 2 hours.

Would that time period also cover any gap in anticoagulation until a P2Y12 agent is fully on board?

Yes, in elective patients it takes 2 hours to get on board with prasugrel or ticagrelor, and it takes 4 hours to get on board with clopidogrel.

Do you think the lower mortality in the STAIR registry’s tirofiban group was due to the reduction in bleeding complications?

Possibly, but there was so little bleeding to begin with. Quite a few patients who died never had a bleeding complication. Certainly the reduction in bleeding might explain why there was reduction in death. There might also be an added clinical benefit with the use of tirofiban that we couldn’t identify by looking at the other events, like myocardial infarction, because those didn’t differ between the two groups (Figure 1). We should also acknowledge that the STAIR registry was not a randomized trial. It was an observational study with multivariable adjusting and propensity matching, by which we did our best to adjust for differences in the baseline characteristics of the patients. There could be selection bias, as in general, people use GP IIb/IIIa receptor antagonists in higher risk patients, but despite that, there was still less mortality with tirofiban. 

Disclosures: Dr. Muhlestein reports no conflicts of interest regarding the content herein.

Dr. J. Brent Muhlestein can be contacted at jbrent.muhlestein@imail.org.

1. Muhlestein JB, Hackett IS, May HT, et al. Safety and efficacy of periprocedural heparin plus a short-term infusion of tirofiban versus bivalirudin monotherapy in patients who underwent percutaneous coronary intervention (from the Intermountain Heart Institute STAIR Observational Registry). Am J Cardiol. 2019 Jun 15; 123(12): 1927-1934. doi: 10.1016/j.amjcard.2019.03.025.
2. Muhlestein JB. Switching from eptifibatide to tirofiban use in ACS: reducing cost while maintaining quality of care. Cath Lab Digest. Mar 2015;23(3): 1-14. Available online at https://www.cathlabdigest.com/article/Switching-Eptifibatide-Tirofiban-Use-ACS-Reducing-Cost-While-Maintaining-Quality-Care. Accessed July 31, 2019.
3. Fung AY, Saw J, Starovoytov A, et al. Abbreviated infusion of eptifibatide after successful coronary intervention: The BRIEF-PCI (Brief Infusion of Eptifibatide Following Percutaneous Coronary Intervention) randomized trial. J Am Coll Cardiol. 2009; 53(10): 837-845.
4. Rebeiz AG, Dery JP, Tsiatis AA, et al; ESPRIT substudy. Optimal duration of eptifibatide infusion in percutaneous coronary intervention (an ESPRIT substudy). Am J Cardiol. 2004 Oct 1; 94(7): 926-929.
5. Muhlestein JB. The movement to short-infusion tirofiban in the cath lab: the next leap in practice change. Cath Lab Digest. March 2018; 26(3): 30-32. Available online at https://www.cathlabdigest.com/article/Movement-Short-Infusion-Tirofiban-Cath-Lab-Next-Leap-Practice-Change. Accessed July 31, 2019.
6. Shahzad A, Kemp I, Mars C, et al. Unfractionated heparin versus bivalirudin in primary percutaneous coronary intervention (HEAT-PPCI): an open-label single centre, randomized trial. Lancet. 2014; 384(9957): 1849-1858.
7. Stone GW, Witzenbichler B, Guagliumi G, et al; HORIZONS-AMI Trial Investigators. Bivalirudin during primary PCI in acute myocardial infarction. N Engl J Med. 2008; 358(21): 2218-2230.
8. Steg PG, van’t Hof A, Hamm CW, et al; EUROMAX Investigators. Bivalirudin started during emergency transport for primary PCI. N Engl J Med. 2013;369(23):2207-2217.
9. Valgimigli M, Frigoli E, Leonardi S, et al; MATRIX Investigators. Bivalirudin or unfractionated heparin in acute coronary syndromes. N Engl J Med. 2015; 373(11): 997-1009.
10. SAfety and Efficacy of Femoral Access vs RadIal Access in STEMI - SAFARI-STEMI. American College of Cardiology: Latest in Cardiology. Presented by Dr. Michel R. Le May at the American College of Cardiology Annual Scientific Session (ACC 2019), New Orleans, LA, March 18, 2019. Available online at https://www.acc.org/latest-in-cardiology/clinical-trials/2019/03/16/23/57/safari-stemi. Accessed July 31, 2019.