The Risks of Switching From VKA Therapy to NOAC Therapy in Frail Older Patients With AF

Older patients with atrial fibrillation (AF) and frailty who switch from treatment with vitamin K antagonists (VKAs) to treatment with non-VKA oral anticoagulants (NOAC) experience more major and clinically relevant nonmajor bleeding complications, according to study findings published in Circulation.

Ambiguity as to whether older patients with AF living with frailty should be switched from VKA to NOAC led the researchers to conduct a superiority trial to determine the more efficacious treatment approach. The primary endpoint was a major or clinically relevant nonmajor (CRNM) bleeding complication, accounting for death as a competing risk. An intention-to-treat analysis was conducted. Thrombotic events were included as secondary endpoints.

The researchers conducted a pragmatic, open-label, multicenter, randomized controlled trial (FRAIL-AF) in the Netherlands from January 2018 through June 2022. This included 1323 patients randomly assigned to either switch from a VKA to a NOAC (n=662) or continue with a VKA (n=661). Included patients were older, living with frailty (age ≥75 years plus a Groningen Frailty Indicator (GFI) score ≥3) and had AF. Patients with valvular AF or with a glomerular filtration rate of less than 30 mL/min/1.73m² were excluded.

All patients were followed-up through 12 months (mean follow-up duration, 344 days), and 90 patients died during follow-up (31 cardiovascular deaths [12 intervention arm; 19 control arm]; 10 bleeding deaths [5 in each arm]). The researchers defined a major bleeding event as bleeding in a critical area or organ, bleeding leading to a decrease in hemoglobin level of at least 2 g/dL, bleeding leading to a transfusion of 2 or more units, or a fatal bleeding. A CRNM bleeding complication was defined as not major bleeding but leading to face-to-face consultation, health care professional medical intervention, or hospitalization.

Among all patients (mean age, 83 years; median GFI, 4), 163 primary endpoint events occurred (101 in the switch arm; 62 in the continue arm), and as per prespecified analysis protocol, the trial was stopped for futility.

The researchers noted the primary endpoint hazard ratio (HR) was 1.69 (95% CI, 1.23-2.32; P =.00112). The thromboembolic events HR was 1.26 (95% CI, 0.60-2.61). Numerically, the intervention arm vs control arm had more urogenital (20 vs 11) and gastrointestinal (17 vs 4) bleedings. HR for the first 100 days was 1.17 (95% CI, 0.70-1.96) in sensitivity analysis and 2.10 (95% CI, 1.40-3.16) for days 100 to 365.

Study limitations include a patient population tolerant to VKA treatment, the choice of NOAC at the discretion of treating physicians, and the underpowered sample size.

“In this pragmatic randomized trial in older patients with atrial fibrillation, living with frailty, more major and/or clinically relevant non-major bleeding complications were observed when switching from vitamin K antagonist treatment to a non-VKA oral anticoagulant, compared to continuing VKA treatment,” the researchers wrote. “This higher bleeding risk with NOACs was not off-set by a reduction in thromboembolic events, albeit the risk of thrombo-embolic events was low in both treatment arms.”

Disclosure: This research was supported by Boehringer-Ingelheim, BMS-Pfizer, Bayer, and Daiichi-Sankyo. Some study authors declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of authors’ disclosures.