When the Time Comes to Stop Oral Anticoagulation: Options and Obstacles

At some point, patient circumstances may warrant stopping these lifesaving drugs, but guidance and data are limited.

When the Time Comes to Stop Oral Anticoagulation: Options and Obstacles

Oral anticoagulants can prevent thrombotic events and save lives in patients with atrial fibrillation (AF) and other conditions, but they don’t in and of themselves make patients feel better. As such, there may come a day when it’s time to set them aside.

Margaret Fang, MD (University of California, San Francisco), sees both sides of the coin in her hospital-based practice: on the one side, the thrombotic events anticoagulation can prevent, such as pulmonary embolism, deep vein thrombosis (DVT), and stroke, but on the other, the complications that come along with treatment like GI bleeding and intracranial hemorrhage. So the question of whether the therapy’s benefits still outweigh the risks for an individual patient comes up a lot, said Fang, medical director of the anticoagulation clinic at her center.

“Obviously we’re helping try to optimize the safety of anticoagulants, but at some point individual patients have circumstances where anticoagulants are perhaps not the right decision,” she added.

What Comes Into Play?

Considerations around stopping anticoagulation in a patient who has an indication for it are always complicated, according to hematologist Rhona Maclean, MBChB (Sheffield Teaching Hospitals NHS Foundation Trust, England). The reason a patient is taking a blood thinner—whether it’s for AF, a mechanical heart valve, venous thromboembolic disease, or some other condition—is important to factor in when thinking about the balance between the expected reduction in thrombotic risk and the heightened risk of bleeding, she said.

Maclean gave the example of a patient being treated for two provoked DVTs—one related to a long flight and the other related to surgery. If there are concerns about bleeding, a physician may be more inclined to discontinue therapy in this patient than in someone with a mechanical heart valve or a high CHA2DS2-VASc score.

The decision generally incorporates a patient’s clinical picture: how their condition is likely to evolve, what impact current therapy is having on them (related to frequent blood draws and dietary restrictions in patients on vitamin K antagonists, for example), and how well controlled they are on treatment, Maclean said. That all has to be weighed against the potential benefit of continuing therapy.

Fang, too, underscored the combination of factors—which also include a patient’s functional status, ability to regularly take and pay for medications, and fear of side effects—that go into the deliberations. “I think this has been recognized for a long time that this concept of shared decision-making and values and preferences has to be done on an individual one-on-one level, because everybody’s individual life circumstances are very different and anticoagulants can be a burden for some,” she said.

Patient Perspective Is Essential

Indeed, all of the physicians interviewed by TCTMD underscored the key role the patient, and their family members and caregivers, should play in weighing the ongoing need for anticoagulation, a process that is often wrapped up in larger discussions about end-of-life care.

“There is a need to reassess the risks and benefits of anticoagulation in A-fib patients continually, and it’s even more important at the end of life,” cardiologist and vascular medicine specialist Geoffrey Barnes, MD (University of Michigan, Ann Arbor), commented. “I think that’s something that needs to be more than just a formal checkbox, but really a consideration at each visit. And as those patients approach end of life, it’s even more important that we have those discussions about what a patient’s values are, what their treatment goals are, and whether the various treatments they’re receiving align with those.”

If stroke prevention remains the primary goal for a patient, then the conversation moves to whether a direct oral anticoagulant or warfarin makes the most sense, Barnes said. If the goal, however, is focused not on avoiding stroke but on spending the next weeks or months with family without having to worry about things like dietary restrictions, routine blood draws, cost, and bleeding/bruising, then maybe it’s time to drop the anticoagulant.

There is a need to reassess the risks and benefits of anticoagulation in A-fib patients continually, and it’s even more important at the end of life. Geoffrey Barnes

Some conversations may be more difficult than others, Maclean said, depending on what has sparked the interest in reconsidering anticoagulation. “For people with injurious bleeds leading to hospital stays, it’s easier to have those discussions and make those decisions,” she said. “I think the very difficult ones are those [with people] who are coming toward the end of life with cancer and other issues or who are getting frail and elderly.”

Overall, Maclean said, “I think the story is [that] we’ve got to have the conversation with patients about what they want at the various stages of their life.”

Referring to the use of oral anticoagulation to prevent strokes in patients with AF, Barnes said “this is probably one of the best cases for shared decision-making in cardiovascular medicine—should you take a medicine that is purely preventive but has side effects or risks and doesn’t make you feel better? That decision has to be a shared decision-making discussion with the patient and often with their family or other caregivers.”

The possibility of preventing stroke through other means, which in the case of AF can mean use of left atrial appendage occlusion, also warrants discussion, Barnes said. That wouldn’t make sense for someone receiving hospice care, though for an older patient who is starting to ask whether oral anticoagulation is worth the hassle, perhaps a device would be a good option, he explained.

“I think that’s something that as a profession we’re going to have to keep sorting out here as left atrial appendage occluders become increasingly utilized and we understand when they’re best suitable for different types of patients,” Barnes said.

Where These Conversations Are Happening

From Maclean’s perspective, the continuing need for anticoagulation is not sufficiently discussed in routine clinical practice. Clinicians generally fear the complications of anticoagulation but they’re also hesitant to stop it once it’s started. In the United Kingdom, patients on oral anticoagulation are subject to annual reviews of safety, but that check doesn’t necessarily include a discussion of whether it’s time to discontinue treatment.

“I think there’s a great concern overall in primary care about doing the wrong thing with anticoagulation,” Maclean said, chalking that up to uncertainty stemming from a lack of knowledge and confidence.

The issue can be bigger than just anticoagulation, she added, alluding to the fact that older patients are often taking several drugs chronically—perhaps statins, antihypertensives, and supplements to stave off osteoporosis. “Overall, the discussions with patients as they age regarding their medication, regarding their healthcare wants and requirements, are not adequately had,” said Maclean. “I think too many patients are left on too many drugs that do not make them feel better.”

I think too many patients are left on too many drugs that do not make them feel better. Rhona Maclean

Barnes said he thinks it’s variable how this is being done out in practice. Palliative medicine specialists are among the best at tackling issues like this, but many primary care clinicians are adept at having these conversations as well, he said. “I think sometimes in cardiology we are so focused on the things our guidelines and our quality metrics tell us we’re supposed to do that we can forget to ask the question: is this appropriate for each individual patient? So while we in cardiology may have a little more expertise about the nuance of medications, specific to say A-fib, often the primary care docs and certainly the palliative care docs are thinking about that sort of holistic patient perspective.”

The fact that oral anticoagulation is managed across various specialties—palliative medicine, cardiology, primary care, geriatrics, and others—can serve as an obstacle to having these conversations, Barnes agreed. Many primary care physicians may be hesitant to stop anticoagulation because the cardiologist is seen to be “in charge” of that medication, he said. “Every clinician should be empowered to do what is best for an individual patient, but furthermore we in medicine have to do a better job of communicating with each other around the time of a visit.”

John Dodson, MD (NYU Langone Health, New York, NY), who specializes in geriatric cardiology, agreed that there’s a need to better coordinate across specialties to improve the lifelong management of anticoagulation, to facilitate discussions about potentially stopping it at a certain point, and to help overcome the therapeutic inertia that keeps patients on an anticoagulant perhaps past the point where it’s likely to do any good.

But there are signs of progress, Dodson said: “I feel like we are kind of moving as a specialty towards doing less in people who are elderly and have disabilities where there’s just no benefit to our medical therapies.”

A Call for More Studies of Deprescribing

One issue perhaps standing in the way of more-routine discussions about the possibility of ditching oral anticoagulation as patients near the end of life is that practice guidelines and other documents from professional societies don’t provide explicit instructions about how to do so. The US guidelines, for instance, state only that “anticoagulant therapy should be individualized on the basis of shared decision-making after discussion of the absolute risks and relative risks of stroke and bleeding, as well as the patient’s values and preferences.”

Dodson said, “I think maybe a little more specificity in what to do with anticoagulation in patients at the end of life would be helpful.”

But he and others noted that there’s little evidence that can be used to guide well-supported recommendations around when and how to start thinking about discontinuing anticoagulation.

Pointing to a study published earlier this year in JAMA Internal Medicine showing that a large proportion of nursing home residents with AF and advanced dementia received an anticoagulant in the last 6 months of life, Fang said, “It just goes to show you that there’s guidelines that say ‘start anticoagulants,’ but not really firm guidelines on how to stop anticoagulants aside from just one-on-one [conversations with] your individual clinician.”

A little more specificity in what to do with anticoagulation in patients at the end of life would be helpful. John Dodson

Fang said a study led by one of her mentees, Sachin Shah, MD (University of California, San Francisco), provides some data to help inform the issue. Published in 2019, the analysis estimates the net clinical benefit of oral anticoagulation beyond age 75 in patients with AF, showing that the benefits become marginal after age 87 for warfarin and after age 92 for apixaban (Eliquis; Bristol-Myers Squibb).

It shows “that we really need personalized and individualized decision-making where if you have 6 months left to live maybe the risks and inconveniences and the burdens of being on an anticoagulant are not worth [dealing with],” Fang said.

What would provide a stronger foundation for discussions around discontinuation of oral anticoagulation would be trials of deprescribing, Fang said, noting that there is a US Deprescribing Research Network. Healthcare systems are not really built to take medications away as patients age, and there is research underway to explore the best ways to do that, she added. “We should really push for deprescribing trials to see if taking medications away and deprescribing them is going to lead to better quality of life and at least equivalent outcomes.”

Fang also she’d like to see more research on patient perspectives regarding bleeding, other complications, and the overall impact of being on oral anticoagulation. Though prior studies have indicated that patients would rather have a bleed than a stroke, that doesn’t reflect what’s seen in clinical practice when older patients develop bleeds and end up needing to be hospitalized, to have procedures, or to get transfused, she said, adding that the threat of bleeding can have a psychological impact on patients.

“I do think that the general push is very strong in favor of anticoagulants, and of course I support prevention of stroke,” Fang stated. “At the same time, it is nuanced topic, and I think we as clinicians all recognize that. And so the more you can do to really raise awareness for the need for deprescribing trials or more broad-based understanding of the consequences of bleeding or being on a chronic medication, I think those are all really important.”

Todd Neale is the Associate News Editor for TCTMD and a Senior Medical Journalist. He got his start in journalism at …

Read Full Bio
  • Barnes reports consulting for Pfizer/Bristol-Myers Squibb and Janssen.
  • Fang reports receiving research support through her institution from the US National Institutes of Health and the Patient-Centered Outcomes Research Institute.
  • Dodson and Maclean report no relevant conflicts of interest.