AFib Patients Often Misjudge Stroke Risk vs Bleeding Risk on Oral Anticoagulants

April 17, 2019

Patients with atrial fibrillation (AF) seem to overestimate their risk for stroke in the absence of oral anticoagulation (OAC) therapy, as well as their risk for serious bleeding complications when on OAC, suggests a questionnaire-based study.

The findings point to a major gap in patients' knowledge about the risks for both AF and OAC that might affect their decisions about whether to go on the stroke-preventive therapy, and "potentially their compliance" once on OAC, write the authors of the study, published online March 29 in Mayo Clinic Proceedings.

"We asked them, what do you think your annual risk of stroke is," Mohamad Alkhouli, MD, West Virginia University School of Medicine, Morgantown, told | Medscape Cardiology.

"We found that over half of the patients thought the annual stroke risk is more than 20%, and that's why they were taking a blood thinner," said Alkhouli, who is senior author on the analysis; the lead author is Mohammad Hijazi, MD, from the same institution.

Of 287 patients responding to the questionnaire, all of whom were aware that AF without OAC is associated with stroke risk, about 53% said they believed that risk for them exceeded 20% per year.

A similar percentage believed their annual bleeding risk on OAC was greater than 10%.

Yet the annual stroke risk was less than 10% for about 90% of the cohort, as estimated by CHA2DS2-VASc scores. Also for 90% of them, the estimated bleeding risk on OAC was 10% or lower per year by HAS-BLED scores; it was no higher than 3% for more than half the study population.

As a result, the authors contend, many of the respondents "may have made an uninformed decision about stroke prevention based on an unrealistically exaggerated perception of their risks."

For better or for worse, Alkhouli noted, it's possible that fewer patients for whom OAC is indicated per guidelines would agree to OAC if they knew their absolute risks with vs without it. On the other hand, "is it okay to be on a blood thinner not knowing the truth?" he proposed.

"I think patients should know, and I was hoping that this would be a bit of an eye opener to cardiologists, that maybe we can be too paternalistic."

Justified Enthusiasm

Physicians "are very enthusiastic about using anticoagulants in AF, and I think that's well justified," Daniel E. Singer, MD, Massachusetts General Hospital, Boston, said in an interview.

"They essentially reverse the risk of stroke posed by AF," said Singer, who isn't associated with the current study. "It's a triumph. And the newer anticoagulants make it safer."

The vast majority of people whom the guidelines say should be on OAC, "by all data, on average, should benefit," he said. "I think if your risk is above, say, 2% per year, the rational decision is to take anticoagulants unless you have an obvious problem with bleeding."

However, he agreed, "the absolute benefit that they're getting, the net benefit, is smaller than what they think it is. It's probably a lot smaller."

Shared decision-making with patients may look different regarding OAC for AF than for some other therapies, observed Singer. Everyone wants to avoid both ischemic strokes and serious bleeding complications. "There isn't that much variability in peoples' preferences, it's really more probabilities and risks that dominate the decision."

But, "shared decision-making in AF has one fundamental problem, which is, it's a balancing of small annual benefits versus small annual harms. And we know that people are not real good with small probabilities," he said.

"The remarkable thing is that OAC has good penetration in practice. A high percentage of people the risk scores say should be on OAC are actually on oral anticoagulation," Singer observed.

"Despite all the different ways of thinking about the problem, lots and lots of people are taking it. And I think for most of them, its probably appropriate for them to take it."

What's the Purpose of OAC?

In the current analysis, shortfalls in patient understanding went even further than the complexities of risk numbers. Only 75.6% of the 173 patients actually on an OAC at time of the study were aware that the treatment was for reducing the likelihood of stroke.

Asked why they were on blood thinners, the remaining patients instead had other answers, such as for lowering the risk for heart attack or simply following doctor's recommendations, the group reported.

"The mere fact that a quarter of our patients didn't even know why they were on a blood thinner is a crime, and these are patients who were cared for by a cardiologist for a long time," Alkhouli said. Some of them may not have understood because of cognitive limitation, he acknowledged, but that's likely only a small number.

"Discussions about risk and benefit need to be individualized, as people's perception of risk varies greatly, and how risk is presented can greatly influence a person's decision to take or not to take OAC," observed Deirdre A. Lane PhD, University of Liverpool, United Kingdom, for | Medscape Cardiology.

Moreover, "treatment decisions about OAC are often undertaken when patients are first diagnosed and often know very little, if anything, about AF," she said by email.

"Some patients will need time to make the decision about OAC, and may need to have more than one conversation with a doctor or healthcare provider to come to a decision."

Oral anticoagulation, Lane said, "is an effective and safe treatment for AF, and patients should be made aware of that. There are very few patients in whom OAC is absolutely contraindicated, and in practice we would always try to encourage patients to take OAC, and revisit the decision at subsequent appointments if they were not on OAC and it was indicated."

She pointed out that in the current analysis, 68.5% of the 54 patients not on OAC cited "worried about side effects," that is serious bleeding, as the reason.

"Giving patients information and the opportunity to ask questions may reduce these concerns and encourage uptake and adherence."

Patient-Perceived vs Estimated Risks

Of the 227 respondents at one tertiary care center who were aware that AF raises stroke risk, about 53% had paroxysmal AF and 47% had persistent or permanent AF.

There were at most "negligible" correlations between patient-perceived and estimated annual risks of stroke in the absence of OAC and of serious bleeding in its presence.

Estimated vs Patient-Perceived Annual Stroke Risk in AF Without OAC
CHA2DS2-VASc Score Range % of Patients With Estimated Stroke Risk in Score Range % of Patients Perceiving Their Stroke Risk in Score Range
1% to 5% 53.7 15.9
6% to 10% 37.0 14.5
11% to 20% 9.3 16.7
21% to 50%* 32.6
>50%* 20.3
*CHA2DS2-VASc scores don't accommodate the range, but it was a possible answer on the questionnaire. They are an index based on whether a patient has congestive heart failure/LVEF ≤40%, hypertension, age ≥75 years, diabetes, stroke/TIA/thromboembolism history, vascular disease, age 65–74 years, or female sex. Mean for cohort, 4.3.

Many patients were close to accurate in their perception of how much OAC can cut the risk of AF-related stroke. As many as 45% believed it would lower it by half, which is close to relative-risk reductions reported in the literature, Alkhouli said.

Others were more optimistic about the efficacy of OAC in stroke prevention. About one-fourth cited a risk reduction of about 70%; another 19% said OAC reduces the risk by 90%.

Estimated vs Patient-Perceived Annual Bleeding Risk in AF While Taking OAC
HAS-BLED Score Range* % of Patients With Estimated Bleeding Risk in Score Range % of Patients Perceiving Their Bleeding Risk in Score Range
1% to 3% 54.0 28.4
4% to 6% 31.6 11.6
7% to 10% 13.5 6.5
>10% 0.9 53.5
*HAS-BLED scores are index based on hypertension, abnormal renal/liver function, stroke, bleeding history or predisposition, labile International Normalized Ratio, age ≥65 years, concomitant drugs or alcohol. Mean for cohort, 2.3.

The cohort was diverse in type of AF, education level, risk profile, history of bleeding, and other characteristics that could potentially influence patient perceptions of stroke and bleeding risk, the authors acknowledge.

For example, about 40% of the cohort had a history of bleeding, with gastrointestinal bleeding cited in about half of cases. A history of bleeding had an odds ratio (OR) of 3.26 (95% CI, 1.27 - 8.37; P = .01) as a predictor of incorrectly perceived risk of bleeding.

Other significant predictors included AF duration longer than a year, being female, and a HAS-BLED score greater than 3, the group reported.

"It's clear in the literature that blood thinners have reduced the risk of stroke, but it's also clear that we don't know how to stratify patients, and we don't give them the information they need to make the decision," said Alkhouli.

He proposed that patients might understand the issues better if physicians framed the risks and benefits differently, including the use of absolute risks along with relative-risk numbers. Patients may not appreciate the difference if they don't hear both. They may also have a different perspective if the risk is expressed as a benefit.

For example, the guidelines would indicate OAC for patients with a CHA2DS2-VASc score of at least 2, which corresponds to a 2.2% annual stroke risk, Alkhouli observed. An impressive-sounding 50% reduction would take that risk to 1.1%, an absolute change that might not impress as much.

Patients would arguably have a fuller appreciation of the risk–benefit balance, he said, if they were also told, for example, that their risk of not having a stroke is 97.8% without OAC and 98.9% with OAC.

"Patients vary in their desire for information," Lane observed. "Not all patients will want or understand absolute risk and relative risk on or off OAC, but some patients will want or like this level of detail. I think it is important to tailor discussions and information to the individual patient."

Hijazi, Alkhouli, and their coauthors report that they have no competing interests. Singer has previously disclosed serving as a consultant or advisor for Boehringer Ingelheim, Bristol-Myers Squibb, CVS Health, Johnson & Johnson, Merck, and Pfizer; and receiving research funding from Bristol-Myers Squibb and Boehringer Ingelheim. Lane as previously said that she has no relevant disclosures.

Mayo Clin Proc. Published online March 29, 2019. Abstract

Follow Steve Stiles on Twitter: @SteveStiles2. For more from | Medscape Cardiology, follow us on Twitter and Facebook.


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.
Post as: