Antithrombotic Therapy: Bleeding Adverse Event Defined

Key opinion leaders discuss the varying definitions of major bleeding as an adverse event in antithrombotic treatments.

Paul Dobesh, PharmD: We always have to look at the other side of the coin because we’ve been in medicine long enough to know that rarely do you get to have your cake and eat it too, right? Typically, if you provide a more potent antithrombotic regimen, the trade-off for that is more major bleeds. So when you look at this, when you look at the study, yes, there is some more major bleeding, but it’s very important for us as clinicians to look at this definition of major bleeding.

It was modified, but it wasn’t modified by the company. You can actually read this in the New England Journal of Medicine paper. It says that regulatory agencies, and we now know that to be the European Union, wanted the definition of bleeding modified. So, it wasn’t something that the company did to mess around with the data, right? It was actually the regulatory agencies that wanted this. Let’s look for a second. You’ve got the ISTH [International Society on Thrombosis and Haemostasis] definition of major bleeding, and then you’ve got this modified. So, what are the similarities, what are the differences? Fatal bleeding—that’s in the ISTH definition; that’s also in this definition. That’s the same, OK? You’ve got bleeding into a critical organ. That’s intercranial hemorrhage, pericardial bleeding, retroperitoneal, things like that. That’s in the ISTH definition; that’s also in this modified definition. You’ve got bleeding that leads to reoperation. That’s in the ISTH; that’s in this definition. So, all 3 of those are the same.

What’s different? The ISTH has bleeding that leads to a hemoglobin drop of at least 2 g/dL, and the regulatory agency said, “’You know, we’re not really that interested in a laboratory-based definition of major bleeds.” I don’t disagree, right? I mean, you can give somebody enough fluids and drop their hemoglobin by 2. So that’s out. That’s in ISTH; that’s not in this one. The other part of it, as well, is bleeding that requires at least 2 U of blood cells, whether the packed red blood cells or other types of blood products. Once again, the regulatory agency says, you know, the need for transfusion is fairly subjective, and I would argue that that’s not wrong. I think in the United States, we’ve become much better about who gets transfused, but it’s still subjective, especially around the world, how much people get transfused and things like that. So that’s out. So, what they put in its place was bleeding requiring, it’s labeled as “bleeding that required hospitalization,” but realize you did not have to be admitted to the hospital. If you visited an ER [emergency room], that counted. If you visited urgent care to get your nose packed because it wouldn’t stop bleeding, that counted. So, it’s really what we call “bleeding requiring medical attention.”

Now, if you’re familiar with that bleeding definition, the ISTH has major bleeding, and then they have what they call clinically relevant nonmajor bleeding. Clinically relevant nonmajor bleeding is bleeding requiring medical attention. So that’s now in this definition.

So, at the end of the day, when you look at this modified definition of major bleeds, it’s higher, like I said, with the low-dose aspirin, adding low-dose rivaroxaban versus aspirin by itself, and it’s significantly higher. But it’s very interesting that when you look at things that are the same between the definitions: so, fatal bleeding, no difference, very small numbers; bleeding into a critical organ, no difference, very small numbers—remember, we’re talking about chronic patients here; bleeding requiring surgery to go in and fix it, no difference, very small numbers. Almost all the difference in bleeds came in this bleeding requiring medical attention, which in every other study is not even a major bleed, right? It’s a clinically relevant nonmajor bleed. So, it’s just very important for us that when we’re trying to look and compare, we’re looking at net clinical benefit type things. You want to make sure you’re comparing things that have equal impact. Bleeding into your brain is a big deal, OK? That would have an equal impact of, say, a CV [cardiovascular] death, MI [myocardial infarction], or a stroke. But going to urgent care is not the same as having an infarct.

So, you just have to make sure that when you’re evaluating—because a lot of people, I think, look at the data and just look at it at the end and say, “Oh, yes, you had this reduction in CV death, MI, and stroke, but you had this increase in major bleeds, and so it’s a wash; don’t bother.” You have to do more critical literature evaluation than that. When you actually look at, really, what we might call major bleeding events, that is much smaller than when you look at this modified definition. So, I think it’s really important for us to encompass the entire, COMPASS, right? It’s very important for us to look at the entire picture here.

So, what we see here is the ability to significantly reduce CV death, MI, and stroke, especially CV death, especially stroke, significantly by adding low-dose rivaroxaban to aspirin, and although there is some more major bleeding, most of this bleeding is not the bleeding that leads to death, it’s not the bleeding that leads to major adverse outcomes, but it is bleeding that would require some type of medical attention.

When I say this is a landmark trial, I really believe when you look at these chronic patients with coronary artery disease and peripheral artery disease, these are data that should make us pivot and think about using these therapies in these patients.

James Groce III, PharmD: Yeah, I agree, Paul. I think one of the things, and you pointed this out nicely, I believe that, I often try to remind our providers here, and certainly when rounding with the internal medicine teaching service as I do here at our hospital, I always remind everyone that when you look in the outcomes of safety for the column for which the patients were just randomized to aspirin therapy—if you know me, as I’m fond of saying—there’s a positive integer into every one of those parameters for which we are evaluating. It’s not as if to reduce the outcomes of efficacy or to improve upon those, that we’re still not seeing the potential for bleeding with just aspirin. So, I think the context, that you put it into in terms of the net clinical benefit—we’re making such tremendous impact upon the outcomes of efficacy for the addition of not only an antiplatelet agent but now an antithrombotic agent, as well, with really no marked increase in the kinds of bleeding that bring, again, death, intercranial hemorrhage, etc, those kinds of things. Those are usually the ones that we care about and certainly that patients care about.

I think the other thing that I would just like to again revisit, comment upon, is the issue of the PAD data, and I know they may not be for some as robust, perhaps. But when you’re that patient for whom you now go back to those very pragmatic issues that you were discussing, about the ability to continue to walk the dog or walk with your wife in the evening or play with the grandchildren or have the garden or just simply walk, those are important things that can be improved upon for this patient population. And then you ask a patient—and, of course, I believe it was statistically significant, but your point was that the end values were low in terms of the patients who required amputation—if you ask a patient simply “Would you like to avoid having an amputation?” I think we know how most of these patients would answer. This would be, again, I think a tremendous benefit for reducing the likelihood of that end point in a patient population, the PAD side of this, that sometimes I worry just simply gets perhaps ignored or we’re doing the best we can, quote, unquote. I think today, certainly, we have something that we can add to the therapeutic armamentarium of those patients, that I believe really improved for that patient subset in preventing the things that bring so much difficulty into their lives as patients who happen to suffer from PAD.