LAAO Tops Oral Anticoagulation for AF in Women and Men: Medicare Analysis

Medicare beneficiaries with atrial fibrillation (AF) see better long-term outcomes when they undergo left atrial appendage occlusion (LAAO) rather than receive oral anticoagulation, a new analysis shows. But the observational study has limitations that may make it difficult to draw firm conclusions.

Through an average follow-up of about a year, risks of mortality and stroke/systemic embolism were lower in both women and men treated with LAAO versus oral anticoagulation, researchers led by Emily Zeitler, MD (Dartmouth Health, Lebanon, NH), report in a study published in the February 14, 2023, issue of Circulation.

Bleeding was increased in the LAAO group early on, but after a 45-day periprocedural period, bleeds were more frequent in the anticoagulation group, regardless of patient sex.

This information can be useful in shared decision-making discussions with patients, particularly women, who have been underrepresented in LAAO trials, study co-author Megan Coylewright, MD (Erlanger Health System, Chattanooga, TN), told TCTMD. She noted that the advantages for LAAO over anticoagulation were seen despite the fact that patients who underwent the procedure were sicker overall, which is consistent with what she sees in clinical practice.

“That’s helpful when I’m talking to my patients—to share that there is a potential for benefit with regards to stroke and bleeding and death in the option of LAAO for those that have an appropriate rationale not to take a long-term blood thinner,” Coylewright said.

I can say it is a reasonable option for those patients who have an elevated risk of bleeding or a history of prior bleeding. Megan Coylewright

The data also provide additional information specifically for women, who have been shown in prior studies to have greater risks of stroke and bleeding related to AF and more complications associated with LAAO compared with men. In 2021, the US Food and Drug Administration said it was looking into potential sex-based difference in LAAO complications.

When speaking with women today, “I share with them that they have two options—both are reasonable, they have different pros and cons—and I listen to what’s most important to the patients,” Coylewright said. “And usually people who are referred to me have a history of fairly severe bleeding or high bleeding risk for which oral anticoagulation is not a good option in the long term. And so I feel this data provides reassurance that there is the possibility for great benefit with this alternative—that is, left atrial appendage occlusion.”

Commenting for TCTMD, Mohamad Alkhouli, MD (Mayo Clinic, Rochester, MN), said the study addresses an important question, but he highlighted several limitations that muddy interpretation of the results.

“Conceptually, I think it is very hard to answer this question from observational data, and even if you do get some hint, it’s mostly hypothesis-generating, so you can’t really be conclusive unless you have a trial,” he said, noting that there are several ongoing RCTs comparing LAAO and oral anticoagulation for stroke prevention in AF.

In the current analysis, Alkhouli pointed to residual confounding and the inability to ensure adequate propensity matching; selection bias stemming from who was offered LAAO and how the index time point was selected for the anticoagulation group; and the lack of a competing-risk analysis.

“It’s hard to come up with any solid conclusions given some of the methodological issues and the inability to adequately control for residual confounders,” Alkhouli said. “We should be patient and wait for the randomized trials with their long-term outcomes that are coming. Observational studies are useful as hypothesis-generating, but we should not overstate their conclusions because they have nonremediable limitations.”

Bolstering LAAO Data Specific to Women

Much of the debate around the efficacy and safety of LAAO revolves around the fact that there have been changes in practice since the initial trials were performed. The early studies of LAAO using the first-generation Watchman device (Boston Scientific) had warfarin as the comparator, but since then, direct oral anticoagulants (DOACs) have taken over as the preferred choice.

Moreover, patients that have been treated with LAAO outside of trial settings tend to be older and to have more comorbidities, and they are more likely to be women, who were underrepresented in the pivotal trials. In the initial Watchman trials, in fact, about 70% of participants were men, and only 224 women were implanted with the device in randomized trials before the technology was approved by the FDA in 2015.

That, along with data showing worse short-term LAAO outcomes among women versus men, provided the inspiration to gather more data specific to women to inform daily decision-making, Coylewright said.

For the current study, Zeitler, Coylewright, and colleagues examined Medicare fee-for-service claims data spanning 2015 to 2019 and used propensity scores to match those treated with LAAO and those treated with oral anticoagulation. After matching, the analysis included 4,085 patient pairs among women (mean age 76; mean CHA₂DS₂-VASc score 5) and 5,378 among men (mean age 75; mean CHA₂DS₂-VASc score 4).

Through follow-up, LAAO was associated with a significant reduction in mortality among women (HR 0.51; 95% CI 0.45-0.58) and men (HR 0.54; 95% CI 0.49-0.60), with similar differences observed for stroke/systemic embolism among women (HR 0.66; 95% CI 0.56-0.77) and men (HR 0.65; 95% CI 0.55-0.76).

Bleeding risk varied over time, however. It was higher in the LAAO group within the initial 45-day periprocedural period, during which the recommended antithrombotic strategy was a combination of anticoagulation and aspirin. After that point, LAAO was associated with lower bleeding risks in both women (HR 0.77; 95% CI 0.68-0.88) and men (HR 0.88; 95% CI 0.78-0.99).

The researchers also included a couple of falsification endpoints—shingles and osteoarthritis—to evaluate the potential for residual confounding. The analysis “failed to detect a difference in groups but does not rule out that one exists,” they write.

Addressing Limitations

The investigators acknowledge that the analysis has some limitations, but point out that “other observational comparative analyses of LAAO and contemporary anticoagulation strategies have demonstrated similar findings, and preliminary data suggest benefits of LAAO over anticoagulation when delivered exclusively with DOACs overall and in sex subgroups.”

They add that “future randomized data are forthcoming to better define this comparative relationship in all comers.”

Coylewright said she’d like to see more discussions of the limitations of randomized trials—namely, their lack of representation of the types of patients seen in everyday practice.

We should be patient and wait for the randomized trials with their long-term outcomes that are coming. Mohamad Alkhouli

“We as a community will need to continue to focus on making sure our research is representative of the patients who we’re seeing in clinic every day,” she said, noting that when there more diversity in the physician workforce and clinical trial leadership, recruitment of women into clinical trials is improved.

Cardiology, however, has had a hard time recruiting and retaining women, Coylewright said. “In part, it’s because in the past . . . we’ve had somewhat of a toxic culture. The work we do to take care of patients and reduce their risk of stroke is difficult work, and the only way that we can move forward together as a field is through respectful and professional interactions to raise the bar.”

As for the issues raised around the current study, Coylewright said, “All of our research has limitations. It’s true. We don’t deny that observational data are limited, and it really is just one piece of information in a broader literature around left atrial appendage occlusion in stroke prevention.”

But it’s a piece that helps fill in some of the gaps in knowledge about LAAO, she said. “There’s a little bit more information specific to women that left atrial appendage occlusion is an option. I wouldn’t say that I can use observational data to definitively say what the magnitude of benefit is, but I can say it is a reasonable option for those patients who have an elevated risk of bleeding or a history of prior bleeding. And I think it’s really important that our patients have choices.”