(195) The use of Gastroprotective Agents in Patients with Atrial Fibrillation Initiating Direct Oral Anticoagulants

Background: Increasing number of studies have shown gastroprotective agents (GPAs) to potentially prevent gastrointestinal bleeding (GIB) in direct oral anticoagulant (DOAC) users. However, the actual utilisation of GPAs in patients with atrial fibrillation (AF) initiating DOACs and the rates of GIB among DOAC users using GPA in clinical practice are not well described.

Objectives: To describe the use of proton pump inhibitors (PPI) at the time of DOAC initiation in patients with AF and their rate of GIB.

Methods: We conducted a descriptive study using data from the Clinical Practice Research Datalink linked to the Hospital Episode Statistics and Office for National Statistics from 2016 to 2022 in the UK. Patients were included if they had an AF diagnosis and initiated a DOAC (apixaban, dabigatran, edoxaban, rivaroxaban) during the study period. Patients who have a history of GIB or other indications for DOACs within 365 days prior to or on DOAC initiation were excluded. The number of patients who used PPI within the first 3 months of initiating DOACs (PPI users) was reported. Patients were followed from the DOAC initiation date until the occurrence of GIB, discontinuation of DOAC (90 days allowable gap), switching to another oral anticoagulant, transfer out of practice, last collection of data, administrative end of follow-up or death, whichever came first. Rates of GIB among DOAC patients with PPI during follow-up were calculated.

Results: Between 2016 and 2022, 25,800 patients initiated edoxaban (PPI users = 10,859, 42%), 4528 dabigatran (PPI users = 1772, 39%), 52,194 rivaroxaban (PPI users = 20,137, 39%) and 109,528 apixaban (PPI users = 47,124, 43%). The mean (standard deviation) age at DOAC initiation ranged from 72 (11) - 75 (11) years and the proportion of females ranged from 38%-45% across the four DOAC groups. The rates of GIB per 100 person-years among PPI users were 2.13 for apixaban, 2.60 for dabigatran, 2.81 for both edoxaban and rivaroxaban. The rates of GIB were higher than the rates of major GIB reported in the landmark trials of DOACs (ranged from 0.47-1.51 per 100 years) which included populations of comparable/younger ages than our study cohort.

Conclusions: In this UK cohort of patients with AF initiating DOACs, GPA use was common with around 40% of patients having concurrent use of PPI. The rate of GIB was higher than those reported in the landmark trials of DOACs. Given that PPIs have been associated with adverse effects and their effects on preventing GIB for certain DOACs, in particular edoxaban, is less clear, further studies are warranted to investigate the impact of GPAs on each DOAC in real-world clinical practice.