GIRAF: Cognitive Outcomes Similar With Dabigatran, Warfarin in Older AF Patients

When it comes to preserving cognitive function in older patients with atrial fibrillation (AF) or flutter, the choice of oral anticoagulant doesn’t appear to matter much, according to the randomized GIRAF trial.

Scores on a variety of cognitive assessments were similar with either dabigatran (Pradaxa; Boehringer Ingelheim) or well-controlled warfarin after 2 years of treatment, Bruno Caramelli, MD, PhD (University of São Paulo, Brazil), reported at the virtual American Heart Association 2021 Scientific Sessions.

Prior research has shown that AF is associated with a greater risk of developing dementia and has suggested that oral anticoagulation can mitigate that risk. As for whether there’s a difference between the direct oral anticoagulants (DOACs) and warfarin, evidence is mixed, although a recently published analysis indicates that dementia and mild cognitive impairment are less likely with DOACs.

GIRAF, the first prospective randomized trial to address the issue, shows that “in older patients with atrial fibrillation or atrial flutter who did not present major cerebrovascular events and were adequately treated with warfarin . . . or dabigatran for 2 years, there was no difference in the majority of the cognitive outcomes,” Caramelli concluded at a media briefing.

Some researchers have proposed that silent cerebral infarcts are driving AF-associated cognitive decline, which means it’s possible that drugs providing more-effective anticoagulation could provide better cognitive outcomes, Jim Cheung, MD (Weill Cornell Medicine, New York, NY), who was not involved in the study, commented to TCTMD. He noted that in the pivotal RE-LY trial versus warfarin, dabigatran (at the higher dose) reduced the risk of stroke or systemic embolism.

But regarding cognitive outcomes, GIRAF “seems to show that—at least when done in a prospective way—maybe there’s not as much of an impact as one would expect” from choosing a DOAC over warfarin, said Cheung, who is a member of the American College of Cardiology’s Electrophysiology Section Leadership Council.

The GIRAF Trial

For the GIRAF trial, conducted in Brazil, Caramelli and colleagues enrolled 200 patients with AF or atrial flutter who were 70 or older and had a CHA₂DS₂-VASc score greater than 1, randomizing them to dabigatran 110 or 150 mg twice daily or warfarin adjusted to an INR of 2 to 3. During the trial, the average time in therapeutic range among warfarin-treated patients was 70%.

Neurologists blinded to randomized group evaluated cognitive function at baseline and 2 years using the following assessments:

• Montreal Cognitive Assessment (MoCA)
• Mini-Mental State Exam
• Neuropsychological battery with a range of tests measuring attention, memory, language, and executive functions
• Computer-generated neuropsychological assessment measuring accuracy and reaction time

The analysis focused on patients who completed a 2-year cognitive evaluation—83 in the dabigatran arm and 66 in the warfarin arm. Caramelli noted that there were more hemorrhagic complications and related deaths with warfarin treatment. Overall, mean age was roughly 75, and 60% of the participants were men.

Baseline scores on three of four cognitive assessments indicated slightly better function among patients in the dabigatran group. After 2 years, there was not much change on any of the assessments, with a statistically significant difference between trial arms seen only for the MoCA (P = 0.02). The difference favored warfarin, although Caramelli suggested it could be due to chance.

When the results were broken down into specific cognitive domains, there were no significant differences between treatment groups in terms of changes in memory, executive function, language, or attention.

Cheung indicated that he wouldn’t read much into the observed difference in MoCA score considering the totality of the findings. “The more tests you do, the more likely you may find a difference with one specific test, but I think taken in aggregate, I would say that it’s basically a wash.”

Uncertain Mechanisms

It’s possible that a larger study might show a difference between the two drugs, Cheung said, “but I think that what it tells me is that the mechanism behind AF-associated cognitive decline is probably multifactorial.”

In addition to silent cerebral infarcts, cerebral hypoperfusion or microbleeds—the latter of which would be worsened with anticoagulation—could also be playing into declines in cognitive function in patients with AF, he suggested. “Fundamentally, what this study shows us is that we still need to better understand what the actual mechanisms are that drive cognitive decline, and that it may be more than just choice of anticoagulation,” Cheung said. “We need to look at whether it’s important to think about rhythm control for certain patients, as opposed to focusing only on the implications of A-fib for causing stroke.”

Caramelli stressed that regardless of which anticoagulant one chooses, it’s important to take it correctly, with a high rate adherence. Otherwise, the results in terms of cognitive outcomes might not be the same as seen in GIRAF.

And when it comes to selecting an oral anticoagulant for older patients with AF, the higher risk of bleeding with warfarin is an important consideration, Caramelli indicated. “If your patient is more prone . . . to suffer a hemorrhagic complication, maybe this is the case to prefer the other drug, dabigatran.”