Share this article on:

Atrial Fibrillation Patients Are Inadequately Treated with Anticoagulants, Study Finds

Fitzgerald, Susan

doi: 10.1097/01.NT.0000516104.16635.57
Back to Top | Article Outline




An overwhelming majority of stroke patients with AF were either taking a sub-therapeutic level of anticoagulants or no anticoagulants, researchers reported. The new study looked at the issue from a different angle by focusing on stroke patients with previously diagnosed AF.

Only about 16 percent of stroke patients with a history of atrial fibrillation (AF) were taking anticoagulant therapy at the time of stroke, according to a study that reviewed more than 94,000 cases.

The overwhelming majority of stroke patients with AF were either taking a sub-therapeutic level of anticoagulants or no anticoagulants, a lapse that could have made them vulnerable to stroke and more severe disability or death.

AF patients who had been taking adequate anticoagulation at the time of stroke were more likely to have a milder stroke and to survive, the study found.

The findings, published in the March 14 online edition of the Journal of the American Medical Association, underscore the long-recognized problem that AF patients are often inadequately treated with preventive medication. The new study looked at the issue from a different angle by focusing on stroke patients with previously diagnosed AF.

“Antithrombotic therapies are known to prevent stroke for patients with atrial fibrillation,” the study noted, and guidelines from the American Heart Association/American Stroke Association and other groups recommend their use. “Despite guideline recommendations, oral anticoagulants such as warfarin are often underused in community practice.”

The study's lead author Ying Xian, MD, PhD, assistant professor of neurology at Duke University, said that if patients in the study had “received the recommended medication they may not have had a stroke in the first place.”

He estimated based on data gathered in the study that between 58,000 to 88,000 strokes could be prevented annually in the United States if AF patients were properly treated with anticoagulants.

Dr. Xian told Neurology Today that while many patients and doctors may shy away from anticoagulants because of fear of bleeding or frequent falls, more attention should be paid in doctor-patient discussions to the risks and benefits of taking these medications.



Lee Schwamm, MD, professor of neurology at Harvard Medical School, who also was a study author, agreed. He said that physicians should be reminded that “recurrent ischemic stroke due to an act of omission (the decision to not prescribe oral anticoagulation) is just as culpable as hemorrhagic stroke due to an act of commission (the decision to prescribe oral anticoagulation).”

It's also important to consider that anticoagulants may help limit the severity of stroke should one occur, making disability or death less likely, Dr. Schwamm added.

“If a clot does form in the presence of anticoagulation, it may be smaller and less well organized, with a greater chance of spontaneous dissolution,” said Dr. Schwamm, the C. Miller Fisher Chair and chief of stroke services at Massachusetts General Hospital.

Back to Top | Article Outline


The retrospective study included 94,474 patients with acute ischemic stroke and known history of AF who were admitted from October 2012 to March 2015 to one of 1,222 US hospitals participating in the American Heart Association/American Stroke Association's Get with the Guidelines program, a national stroke registry and quality improvement program.

Patients were randomized to one of five categories based on the medication they were on seven days before stroke: no antithrombotic therapy; antiplatelet therapy only (aspirin, clopidogrel or both); subtherapeutic warfarin with an international normalized ratio (INR) less than 2 at admission; therapeutic warfarin with an INR of 2 or higher; or a novel oral anticoagulant (NOAC) — edoxaban was excluded from the list because it was only approved in January 2015.

The primary outcome was stroke severity as measured by the National Institutes of Health Stroke Scale (ranging from 0 to 42, with a score of 16 or above indicating moderate or severe stroke). The study also considered in-hospital mortality and functional outcome at discharge as measured by the Modified Rankin Scale.

Prior to stroke, 30.3 percent were not on antithrombotic therapy; 40 percent were taking aspirin or clopidogrel or both, even though they were considered at high risk for stroke; and 13.5 percent were on warfarin, but at a subtherapeutic dose.

The study found that patients taking a therapeutic dose of warfarin or a NOAC were significantly less likely to have a moderate or severe stroke (15.5 percent and 17.5 percent, respectively) than those on no antithrombotic therapy (27.1 percent) or antiplatelet therapy (24.8 percent).

In-hospital mortality was about 21 to 25 percent lower in the group that received the recommended medication prior to stroke compared to the others. Those getting the recommended treatment also were 43 to 45 percent more likely to function independently after stroke, Dr. Xian said.

Among the documented reasons given for not prescribing medication were risk of bleeding (16.3 percent), risk of falls (10.3 percent), terminal illness (6.2 percent), patient or family refusal (4.3 percent), mental status (1.1 percent), medication adverse effects (1 percent), or allergy (0.6 percent).

However, 65 percent of the time there was no documented reason, though that doesn't mean there wasn't a valid one. On discharge, many of the patients still were not on anticoagulants, even though the drugs are recommended for secondary stroke prevention.

The study's authors acknowledged several limitations with the study, including the fact that it was a retrospective observational analysis.

Back to Top | Article Outline


Commenting on the study, Larry B. Goldstein, MD, FAAN, FAHA, Ruth L. Works Professor and chairman of neurology at the University of Kentucky, said: “Because the study is limited to those who had a stroke, we don't know how many of those on anticoagulation didn't have a stroke over the same period of observation.”

The idea of using warfarin can be off-putting and even scary to patients and families, he noted. The risk of bleeding varies with different factors; patients must have periodic blood tests to make sure the drug is properly titrated, and there are dietary restrictions.

Dr. Goldstein said it was hoped that the introduction of NOACs, which seem to be at least as efficacious as warfarin with a relatively lower risk of bleeding complications and no requirement for monitoring, would lead to more AF patients being anticoagulated. But he said that this study suggests that many patients who would benefit from treatment are still not receiving either warfarin or one of these drugs.

Although some analyses indicate that the NOACs can be cost-effective from a societal standpoint despite higher costs, they may not be covered by insurance plans, Dr. Goldstein said. “The out-of-pocket costs for patients can be substantial,” he said, adding: “Even though cost may be an important issue, it may not fully explain why so many people with AF still are not on any recommended preventive medication.”

Bradford Worrall, MD, FAAN, Harrison Distinguished Teaching professor and vice chair for clinical research at the University of Virginia, said he was not surprised by the findings. “As a stroke neurologist, I see a lot of people with atrial fibrillation who come in with stroke and are not on anticoagulants or are not being properly treated with anticoagulants,” he said.

Dr. Worrall said that among factors likely to explain that gap include the fact that “physicians and other practitioners may be the ones choosing not to do anticoagulation.”

He said doctors need to dig deeper when a patient presents with a reason for not being on anticoagulants, such as a past episode of gastrointestinal bleeding. Among questions they should ask: How along ago did it occur? Was it minor or major? Also, the fact that a patient is frail and living alone is not a reason to skip anticoagulation therapy, he said. The risk of bleeding from a fall is reasonable to consider, but “it is important to recognize that the decision not to do anticoagulation can have real consequences too,” Dr. Worrall said. If someone's mother “lives alone and is struck by an ischemic stroke, what's the chances of her receiving heroic acute intervention that is going to reverse that process?”

Dr. Worrall said he likes to use an online tool called SPARC ( that estimates an AF patient's risk of stroke and the risks and benefits of antithrombotic therapy based on the person's individual profile. Having concrete risk numbers to add to the discussion with patients is helpful, he said, noting that the decision not to do anticoagulation therapy needs to be an “active decision. It shouldn't be a passive decision.”

Most people with AF are treated by a primary care physician or cardiologist, and many do not come to the attention of a neurologist unless they have a stroke, noted Frederick A. Masoudi, MD, MSPH, FACC, FAHA, a cardiologist and professor of medicine at the University of Colorado Anschutz medical campus.

There are several reasons why anticoagulants may not be prescribed, he pointed out. “The physician may consider the patient's risk of bleeding too high to counterbalance the beneficial effects of anticoagulation,” Dr. Masoudi said, pointing out that many of the factors that predispose patients to stroke in AF also predispose to bleeding. “However, it is not always clear that clinicians appropriately balance the potential risks and benefits based upon data. In these situations, clinicians and patients would benefit from shared decision-making tools supported by evidence-based assessments of the risks and benefits of therapy that could be deployed to support better decision-making.”

Sometimes, patients may not be adherent to the medications that are prescribed for different reasons, he said. Educating patients on the rationale for the use of anticoagulants and exploring the reasons for non-adherence may be helpful (discussing, for example, the affordability of one medication over another).

Finally, Dr. Masoudi said, clinicians may just not be prescribing anticoagulants to those who, based upon a risk-benefit analysis, would derive benefits from therapy. This is an area where systematic interventions to improve quality of care by identifying these gaps and working with clinicians to improve the practice are helpful.

Anticoagulation for AF in ambulatory practice is a focus of the American College of Cardiology PINNACLE registry program, which reports treatment rates with national benchmarks to practices, Dr. Masoudi said, allowing them to improve their adherence to this important process of care.

Back to Top | Article Outline






Back to Top | Article Outline


•. Xian Y, O'Brien EC, Liang L, et al Association of preceding antithrombotic treatment with acute ischemic stroke severity and in-hospital outcomes among patients with atrial fibrillation JAMA 2017; Epub 2017 Mar 14.
    © 2017 American Academy of Neurology