Direct Oral Anticoagulation for Post-TAVI Atrial Fibrillation

Mr. Gannon is an 82-year-old man with severe symptomatic aortic stenosis, hypertension, and benign prostatic hyperplasia (BPH). He was admitted to the hospital and underwent successful transcatheter aortic valve implantation (TAVI). However, he developed postprocedural atrial fibrillation and was started on metoprolol for rate-control management. The decision was made to initiate anticoagulation with warfarin. However, interns on the team asked whether direct oral anticoagulants (DOACs) would be an appropriate substitute to reduce risk of thromboembolism.

The  prevalence of aortic stenosis in patients older than 75 years is about 12% and the prevalence of severe aortic stenosis is 3%. Complications of aortic stenosis include heart failure, angina, syncope, and atrial fibrillation. For those with symptomatic aortic stenosis or asymptomatic severe aortic stenosis, aortic valve replacement (AVR) can reduce AVR can be achieved surgically (surgical AVR or SAVR) or via transcatheter AVR, or more accurately referred to as transcatheter aortic valve implantation (TAVI) because the diseased native aortic value is not actually excised.

The choice of TAVI versus SAVR is based on individual patient characteristics including comorbidities, age, expected length of survival, vascular access, and patient preferences. But for patients with high or prohibitive surgical risk, TAVI can reduce mortality, symptomatic aortic stenosis, and hospitalization. Postoperative complications associated with TAVI include stroke, paravalvular leak, stent migration, conduction abnormalities, and iliofemoral complications (e.g., hematoma or pseudoaneurysm).

Recommendations for antithrombotic prophylaxis after TAVI include aspirin, clopidogrel, or warfarin for at least 3 months. The use of DOACs in patients with mechanical heart valves is contraindicated. However, the off-label use of DOACs in patients with bioprosthetic heart valves is increasingly common. Studies that compare the use of DOACs such as rivaroxaban to warfarin in patients with bioprosthetic mitral valve and atrial fibrillation report similar mortality, major cardiovascular events, or major bleeding events. Studies that compare rivaroxaban to antiplatelet therapy in adults without indication for oral anticoagulation after TAVI report increased mortality with rivaroxaban.

In the ENVISAGE-TAVI AF trial, investigators randomized nearly 1400 individuals with severe aortic stenosis after TAVI and atrial fibrillation to receive edoxaban or warfarin. Nearly all patients (99%) had atrial fibrillation prior to TAVI or developed post-procedural atrial fibrillation (1%). After a mean follow up of 1.5 years, edoxaban was found to be noninferior to warfarin for the composite primary efficacy outcome of death from any cause, myocardial infarction, ischemic stroke, systemic thromboembolism, valve thrombosis, or major bleeding (17.3 vs. 16.5 per 100 person-years; 95% CI, 0.85 to 1.31; P=0.01 for noninferiority).

These results suggest that in patients such as the one described above with atrial fibrillation following TAVI with a bioprosthetic valve, edoxaban may be a reasonable alternative to warfarin.

The following NEJM Journal Watch Summary provides further details of the study.

Mark S. Link, MD, reviewing Van Mieghem NM et al. N Engl J Med 2021 Aug 28

Patients with atrial fibrillation (AF) undergoing transcatheter aortic valve replacement (TAVR) traditionally receive vitamin K antagonists (VKAs) because of the long history of warfarin use in patients with valve replacements and the inferiority of direct-acting oral anticoagulants (DOACs) compared with warfarin in those with mechanical valves. In the industry-funded, multicenter, open-label ENVISAGE-TAVI AF trial (NCT02943785), researchers compared the safety and efficacy of the DOAC edoxaban and a VKA in this setting.

Among 1426 patients randomized to edoxaban or a VKA for prevalent (99%) or incident AF after successful TAVR, during a median follow-up of approximately 1.5 years, the groups had similar rates of the composite primary efficacy outcome of all-cause death, myocardial infarction, ischemic stroke, systemic thromboembolism, valve thrombosis, or major bleeding (17.3 and 16.5 per 100 person-years for edoxaban and VKA groups, respectively; P=0.01 for noninferiority of edoxaban). The rate of major bleeding was significantly higher with edoxaban compared with VKA (9.7 vs. 7.0 per 100 patient-years; hazard ratio, 1.40). The excess bleeding risk in the edoxaban group was largely attributable to GI bleeding.

Comment: These trial findings are important given the increasing use of TAVR around the world and concerns regarding use of DOACs in these patients. Whether these results in edoxaban can be extended to other DOACs is unclear. However, this trial opens the door for use of DOACs, and especially edoxaban, in patients undergoing TAVR.

Browse more From Pages to Practice »

James Yeh, MD MPH is an internist and assistant in medicine at the Massachusetts General Hospital and an instructor in medicine at Harvard Medical School. His clinical and academic interests are in evidence-based medicine, medical education, continuing medical education, cardiopulmonary diseases, cardiovascular risk reduction, critical illness, care transition, polypharmacy, and health communication. He was a NEJM editorial fellow in 2015-2016.