Reversal Agents for DOACs: When Should You Use Them?

Elaine M. Hylek, MD, MPH; Charles V. Pollack, Jr, MD, FACC; Christian T. Ruff, MD, MPH


January 22, 2019

Editorial Collaboration

Medscape &

Christian T. Ruff, MD, MPH: Hello. My name is Dr Christian Ruff, and I am the director of general cardiology at Brigham and Women's Hospital and Harvard Medical School. I'm delighted today to be joined by two colleagues and close friends: Dr Elaine Hylek, who is professor of medicine at Boston University; and Dr Charles Pollack, who is professor of emergency medicine at Thomas Jefferson University. Welcome to both of you.

Charles V. Pollack, Jr, MD, FACC: Thanks, Christian.

Elaine M. Hylek, MD, MPH: Thank you.

Ruff: Today we are going to discuss a really important topic: management of anticoagulation-associated bleeding in the era of specific reversal agents. We know that novel oral anticoagulants (NOACs) are now broadly used as anticoagulants for stroke prevention and atrial fibrillation (AF). We will talk a little bit about what an important advance this is, but obviously any time you prescribe a patient a blood thinner, it can increase the risk of bleeding.

We will talk about how common this bleeding is and how we would manage bleeding in general. Then we will end by focusing on the role of this exciting advance that we have in antidotes, or specific reversal agents, for both types of NOACs.

Elaine, we will start with you. You have obviously been in this field for a long time, since back in the days of warfarin. In the current age, comparing warfarin and the NOACs, how commonly do we see bleeding in AF patients in general?

Bleeding in AF: Warfarin vs NOACs

Hylek: I think it's important to start with the International Society on Thrombosis and Haemostasis definition of [major] bleeding, which includes fatal, critical organ bleeding, like intracranial hemorrhage (ICH), as well as any acute bleed that is associated with a 2-g/dL drop in hemoglobin or 2 units or more of packed-cell transfusion.[1] This trial used the same definitions. If we just start there with our AF as well as venous thromboembolism patients, the incidence of major bleeding is about 3% in a trial population per year. I believe that ICH is about 5 per 1000, or about 0.5%, on warfarin. It's a devastating complication, but thank goodness it's relatively rare.

With good renal function, we should not get into trouble that often.

If we think about warfarin, which we all have used for our whole careers, it's not necessarily easy to hurry up and reverse that effect. Remember that warfarin's half-life in the older population is about 60 hours. Even though we give vitamin K and can certainly use the prothrombin complex concentrates, it's tough because warfarin's half-life is so long.

One of the biggest advances of these NOACs or DOACs (direct oral anticoagulants) is just the fact that they have a short half-life—on average, I think about 10 hours. We know from clinical trials that fatal bleeding, certainly ICH, was reduced by 50%.[2,3,4] That is a huge plus. We always have the short half-life working in our favor. With good renal function, we should not get into trouble that often.

Ruff: Elaine, I think you mentioned the most important message, that NOACs are vastly safer without a reversal agent. They were all studied before the advent of reversal agents.

Obviously, being in cardiology and medicine, Elaine and I prescribe these drugs like water. Charlie, you are in the emergency department (ED) and you see bleeds in real time. Do you see a lot of NOAC-associated bleeding? What is the severity in the broad swath of patients that you care for?

Pollack: It's interesting, Christian. Even though we are in an era where more and more new starts of anticoagulation are NOACs, there are still a lot of patients on warfarin. Obviously, patients we see in the ED are not a randomized population. These are people that come because they think they are having a problem. Not necessarily do we see outright hemorrhage, but we certainly see more bleeding concerns related to warfarin than we do to NOACs. I think that is consistent with the overall message that these drugs are safer than warfarin. They are certainly easier to manage.

One advantage of warfarin in the ED is that we can measure how anticoagulated somebody is. That, of course, can't be applied to the NOACs, but fortunately we don't need to know very often how anticoagulated these NOAC-treated patients are.

I can tell you that in terms of severe bleeding, particularly ICH—and Elaine mentioned the numbers in the clinical trials and then in postmarketing surveillance as well—it looks like the risk for ICH on NOAC is a good 50%-60% lower than it is on warfarin. That has certainly been my experience in an urban quaternary-care ED. I've seen, I think, one ICH on a NOAC in a couple of years and at least a half dozen or more on warfarin. Now, we don't know what the denominators are, but certainly it's more common to see these catastrophic bleeding problems on warfarin than it is on any of the NOACs.

Common-Sense Measures to Reduce Bleeding Risk

Ruff: Elaine, you have spent a long time caring for these patients longitudinally. The best way to manage a bleed is to prevent it from occurring in the first place. Are there common-sense things we can tell providers and patients—generally older populations who are going to be chronically anticoagulated—to do to avoid having a bleed in the first place? We have seen in clinical trials, also in real-world evidence, that a lot of patients have an unnecessarily high risk that is modifiable.

The other thing that I think is absolutely critical is reading the package insert.

Hylek: Absolutely. The field has moved to question how often physicians are also prescribing aspirin in patients already taking an anticoagulant. Globally, there are guidelines now that are really putting up the red flag about combinations of aspirin unless absolutely necessary. We won't get into the percutaneous coronary intervention and acute coronary syndrome guidelines, but even those are trying to minimize dual therapy of an anticoagulant with an antiplatelet.

When I'm teaching my residents, I say, "You need to go through a checklist. What might increase the likelihood of a bleed in this patient?" Use of nonsteroidal, anti-inflammatory drugs (NSAIDs) in the older population is absolutely a contraindication. It really increases the risk for upper gastrointestinal (GI) bleeding. Even though physicians are often hesitant [to prescribe anticoagulants] in the oldest patients because of fall risk—and clearly we all have geriatric patients and fall risk is a concern—I don't withhold therapy because I personally would not want to have an ischemic stroke. I would rather have my visiting nurse come in and make sure that my loose rugs are gone, that I'm wearing the proper footing [shoes], and that I'm not getting out of bed at 3 AM without a bright nightlight on so that I don't slip and fall.

I take bleeding very seriously as an internist. My goal is to keep patients on these medicines without having an untoward event. Certain things, like blood pressure control, are basic elements of care of patients. Poor blood pressure control increases the risk for both ischemic and hemorrhagic stroke. I think those are common-sense things that we should all be doing.

Pollack: I don't often say this because it bores people to death, but the other thing that I think is absolutely critical is reading the package insert. All of the NOACs have specific dosing adjustments to be made, primarily based on renal function, but also on a few concomitant medications. Remember that the concern for drug-drug interactions is much lower with the NOACs than it is with warfarin, but it's not zero. The fact of the matter is that package inserts for the NOACs are very helpful in terms of giving appropriate guidance for dose adjustment. For the most part, that is driven by renal function.

Most people who are taking NOACs for an extended period of time have AF, multiple comorbidities, and tend to be older; and given that renal function tends to deteriorate with age, there are very basic adjustments that can be made in dosing. Again, it goes drug by drug. Sometimes you look at body weight; sometimes you look at age. But with all of them, you look at renal function.

Make sure the patient is on the right dose, which means not only checking creatinine clearance every 6 months or so but also being aware of the fact that acute intercurrent illness can cause significant temporary alterations in renal function. Patients can suddenly be over-anticoagulated from bioaccumulation of NOACs.

Although we recognize that these drugs are a lot safer than warfarin, we do have to be careful with them and make sure that people are on the right dose.

Ruff: That's a great point. People are always wanting to measure something with these drugs because there isn't an INR equivalent. Like you, I tell people that if they want to measure something, measure creatinine clearance; because if they measure that one thing, they are always going to be able to provide either the appropriate full dose or reduced dose.

Managing Bleeding Associated With NOACs

Ruff: Despite our best measures in avoiding antiplatelet therapy and unnecessary NSAIDs, and measuring renal function for appropriate dose adjustment, unfortunately we still will encounter bleeds just given the number of patients that we are treating with these drugs.

In general, I hear from both of you that the majority of bleeds we are going to see are the non-life-threatening bleeds. Charlie, In the ED, what is your approach to managing bleeding? What are the general principles and supportive measures for all-comer bleeds? Then we will discuss life-threatening bleeds, which are relatively uncommon but have a very specific management protocol.

Pollack: There are basic measures that you take, even on patients who are not anticoagulated. In patients who come in with a bleeding issue, whether it's severe or not, the first thing you are going to do is try to figure out what anticoagulant they are taking and stop it. You want to make sure they are not getting their next dose. That sometimes involves reminding family members. At least in our ED, it's not uncommon for family members to bring in the bag of meds. When it's time for mom or dad to take their medicine, they will pull it out and give it to them because they don't want the hospital to charge them for it. You have to be careful about that.

Then we obviously start with the ABCs (airway, breathing, circulation). Again, most of the bleeds associated with NOACs are not going to require acute resuscitation or airway management, but we will always start there to make sure that patients are stable.

If there is ongoing bleeding, we try to stop it in any number of ways. If it's at a compressible point, obviously we just apply compression or pressure. If the patient is going to need surgical intervention, we start arranging for that very quickly. We attend to blood pressure support. If patients are losing their blood pressure, we will start first with fluid resuscitation and then move on to blood products if needed. Generally speaking, you are trying to temporize the patient and get the bleeding under control.

When there is an anticoagulant involved and the bleeding is life-threatening or it's in a critical space, then you start thinking sooner instead of later about the potential for reversing the anticoagulant activity. That is a really small proportion of patients that come in with bleeding complications.

Reversal Agents for NOACs

Ruff: Thank you, Charlie. That is helpful. Elaine, we now have two approved reversal agents for the NOACs. Can you briefly tell us about these agents?

Hylek: With dabigatran, which is our direct thrombin inhibitor, there is a monoclonal antibody called idarucizumab (Praxbind). This was studied in the RE-VERSE AD study.[5] They looked at patients who had major, severe bleeding or, importantly, the need for an emergent procedure or urgent surgery. I think those two populations have different needs.

It is also very important to know how these reversal agents work. [The onset of] idarucizumab is very quick. It's two injections of 2.5 g (5 g in total). It has an affinity for dabigatran, 300 times the affinity for thrombin. The study was impressive. It reduced the length of time of the bleeding, as inaccurate a measure as that might be. It still looked very favorable. It's a great addition to our armamentarium for dabigatran-related bleeds.

The factor Xa inhibitors, apixaban and rivaroxaban, have their own specific reversal agent, andexanet alfa (Andexxa). We are still waiting for some data on the emergent reversal for these drugs for the patient who comes in with trauma or needs urgent surgery, but they did indeed study patients with major bleeding, severe bleeding.[6] What we do here is [administer] a bolus, followed by a 2-hour infusion. There are some different doses, based on the initial dose of rivaroxaban or apixaban as well as timing of the last dose of the drug, 8 hours being a cut-off.

We will watch and see. Hopefully physicians are realizing that the cost of these agents may be prohibitive, and that we need to be very careful how we use these agents for instances in which it's absolutely mandatory to immediately try to reverse the anticoagulant effect.

These are really good drugs that you don't use very often.

Ruff: Thank you, Elaine. Charlie, you led RE-VERSE AD with idarucizumab. You have seen a lot of these bleeds and have a lot of management experience. In your practice, which patients are you using these reversal agents in? When do you pull the trigger and prescribe one of these reversal agents? What types of bleeds are you seeing? Other than ICH, what conditions do you use these agents in?

Pollack: The guiding principle is that these are really good drugs that you don't use very often. As we have discussed, these life-threatening bleeds, these bleeds in the critical spaces, are quite rare with NOACs. I agree with you that with an ICH in a patient who is deemed salvageable who is taking dabigatran, you would use idarucizumab. If the patient is taking apixaban or rivaroxaban, you would use andexanet alfa. Those are givens.

If the patient is deemed not to be salvageable, you probably would not want to go through the expense and time of giving these drugs. Otherwise, we are looking at the situation where we need reversal of the anticoagulation in order to safely, efficiently, and effectively manage the patient.

Remember that idarucizumab and andexanet alfa are not hemostatic agents. They don't plug holes. What they do is reverse the anticoagulant for which they are engineered to treat. They basically remove the iatrogenic coagulopathy that the patient is facing. That makes management of the patient noticeably easier for the treating physician.

If you have a bleed that is a nuisance and it can be controlled with local pressure or the patient is stabilized very quickly with a liter of saline, that is probably not a patient you'd want to reverse. If you have a patient who is anticoagulated and those basic measures that we discussed are not effective, you are going to use a reversal drug if the patient continues to bleed, becomes hemodynamically unstable, needs to be rushed to surgery (in the case of dabigatran), or needs to go to surgery if he's not bleeding now but it's deemed unwise to do that surgery without reversing first.

The same thing applies with andexanet alfa, although as Elaine pointed out, in the ANNEXA-4 study,[6] it was looked at only in cases of severe hemorrhage, not in the preprocedural settings.

These drugs are not at all competitive with each other because idarucizumab is only used for dabigatran and andexanet alfa is only used for rivaroxaban or apixaban, but they tend to be lumped together as NOAC reversal agents. They do have different properties. They are dosed in different ways. It's important that acute care, urgent care, and intensive care physicians be familiar with these drugs. For the most part, no offense, but people like you guys who are prescribing the index anticoagulants are generally not going to be faced with a situation where reversal is needed. This is really something that is going to be done in the ED, operating room, or in the intensive care unit.

Ruff: You make a great point. Certainly, removing the anticoagulant from the situation is important, but these patients who are coming in with potentially life-threatening bleeds usually have something else wrong with them.

The mortality rates, even in patients who get a reversal agent, are still high, which reflects the underlying condition that these patients have. It's important that this is a bridge to further management. They are not miracle drugs; although it is remarkable in the studies that they essentially removed the anticoagulant effect within several minutes and have not demonstrated any off-target or prothrombotic effects. Unlike warfarin, these are truly reversal agents, which is an important advance.

Restarting Anticoagulation

Hylek: What do we do after we have controlled the bleeding? With the older AF population, most of these bleeds are in the GI tract. It's important to speak to the endoscopist and ask how soon we can restart this anticoagulant.

Even though this is a scary thing for patients and families, we do know that when these individuals are off their anticoagulant, that is when you see the mortality and the stroke rates increase. Sometimes we are afraid to offer patients to go back on these drugs. I think that is something we should be more aggressive about.

Ruff: When we get them back from Charlie after he has managed our bleeds, we have to do our part and restart. That is important. We know that the majority of bleeds we are going to see are GI bleeds. You make a critical point that the majority of those patients can be safely restarted and don't rebleed, with dramatic reductions in their risk for long-term cardiovascular events and even death.

Having one bleed does not mean that you can no longer be challenged with an anticoagulant. Unfortunately, the way we practice, that frequently is the case.

Pollack: I was one of the authors of the American College of Cardiology pathway for managing bleeding events in patients taking oral anticoagulants.[7] There were 10 or 12 of us writing the paper. We had remarkable unity about how to write every piece of this except for the point Elaine just brought back up, which is, when do you restart anticoagulation? Everybody has their own recipe for doing that.

There is this balancing act between acknowledging what has just happened, which obviously in an intracranial bleed is different from a GI bleed, and balancing that with the risk of the thrombotic event for which the patient was taking an anticoagulant in the first place. There is guidance out there, but it's not quite as consistent as the guidance on to when to start anticoagulation and in those rare cases when it's needed to have to reverse anticoagulation.

Ruff: There have been no real trials to inform us, so it's really the art of medicine more than anything else.

Pollack: They still need us. Robots can't do everything.

Ruff: That's right. I'd like to thank both of you. This has been a terrific discussion. I've learned a lot, as I always do when I talk to the two of you. I hope that our audience has enjoyed the session.

This is truly a remarkable time because we have more effective and safer alternatives to warfarin. Even in those rare cases where we do run into trouble with life-threatening bleeding, we have two very effective and safe specific reversal agents. We will rarely need them, but when we do, they will be in our clinical armamentarium. The psychological reassurance is probably their greatest contribution to clinical medicine.

I want to thank both of you, Elaine and Charlie, and I look forward to seeing you both soon.

Pollack: Thank you.

Hylek: Thank you.


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.