Treating atrial high-rate episodes with edoxaban not effective, may be harmful
Click Here to Manage Email Alerts
Key takeaways:
- Edoxaban did not improve ischemic outcomes and raised bleeding risk in patients with atrial high-rate episodes.
- Anticoagulation should not be given to these patients if they do not have atrial fibrillation.
In patients with device-detected atrial high-rate episodes, treatment with edoxaban did not reduce ischemic events and increased risk for a composite of death and major bleeding compared with placebo, researchers reported.
Atrial high-rate episodes (AHREs) are rare and brief atrial arrhythmias that are sometimes detected in patients with implanted cardiac devices. It was not known whether anticoagulants should be initiated in patients with AHREs but not atrial fibrillation, Paulus Kirchhof, MD, professor of cardiovascular medicine and director of the department of cardiology at University Heart and Vascular Center Hamburg in Germany and professor of cardiovascular medicine at the University of Birmingham in the United Kingdom, said during a press conference at the European Society of Cardiology Congress.
Image: Adobe Stock
“AHREs are detected in every fifth patient with a pacemaker, defibrillator or loop recorder, and are found in up to 30% of elderly people with cardiovascular diseases,” he said. “AHREs look like atrial fibrillation, therefore clinicians often prescribe blood thinners in patients with AHREs based on their effectiveness in patients with atrial fibrillation.”
For the NOAH-AFNET 6 trial, Kirchhof and colleagues randomly assigned 2,536 patients (mean age, 78 years; 37.4% women) with device-detected AHREs lasting at least 6 minutes (median duration, 2.8 hours) and at least one additional risk factor for stroke to edoxaban (Savaysa, Daiichi Sankyo/Eli Lilly) or placebo.
The trial, which was simultaneously published in The New England Journal of Medicine, was stopped early after a median follow-up period of 21 months due to safety concerns and futility, Kirchhof said during the press conference.
The primary efficacy outcome of CV death, stroke or systemic embolism occurred at a rate of 3.2% per patient-year in the edoxaban group and 4% per patient-year in the placebo group (HR = 0.81; 95% CI, 0.6-1.08; P = .15), according to the researchers.
The rate of stroke was approximately 1% per patient-year in both groups.
“What was surprising to us was the very low number of strokes,” Kirchhof said during the press conference. “Systemic emboli and cardiovascular death occurred at approximately the expected rate.”
The safety outcome of death or major bleeding occurred more often in the edoxaban group than in the placebo group (5.9% per patient-year vs. 4.5% per patient-year; HR = 1.31; 95% CI, 1.02-1.67; P = .03), Kirchhof and colleagues found.
“The difference was mainly driven by doubling of major bleeding” in the edoxaban group, Kirchhof said. “This is an expected effect of any anticoagulant, including edoxaban.”
In the total cohort, 18.2% of patients developed AF, for a rate of 8.7% per patient-year, according to the researchers.
“This trial demonstrates that anticoagulation should not be used in patients with atrial high-rate episodes until atrial fibrillation is documented by ECG,” Kirchhof said during the press conference. “That is a practice-changing finding, at least for many clinicians.”
Reference:
Perspective
Back to Top
This was an important and very clinically relevant topic to investigate. We see these AHR episodes all the time in our patients with devices, and do not know whether to treat these episodes or not, and what burden should prompt treatment, etc. So, the medical community has been awaiting the results of this trial.
The finding regarding the higher risk for bleeding with edoxaban is not surprising; the finding that is most surprising is the very low risk for stroke in both arms of the study, especially given the median age of 78 years and having other risk factors for stroke.
We still do not know how this population should be treated. I do not think this trial has definitively answered the question, and I worry that the required duration of the AHR episodes of at least 6 minutes may have biased the results toward neutrality. Patients with much longer AHR episodes are likely not well represented here (median duration of AHR episodes of 2.8 hours). Clinically, I worry far more about patients with longer AHR episodes (8 to 10 hours or more) than patients with AHR episodes lasting for 6 to 10 minutes.
I look forward to the results of the ARTESIA trial, which is addressing the same question.
Collapse
Kirchhof P, et al. Hot line 1. Presented at: European Society of Cardiology Congress; Aug. 25-28, 2023; Amsterdam (hybrid meeting).
Click Here to Manage Email Alerts