Gastrointestinal bleeding among oral anticoagulant users: a comprehensive 7-year retrospective review using Türkiye's national health data system.

Abstract

Background/aim: The comparative risk of gastrointestinal bleeding (GIB) among users of direct-acting oral anticoagulants (DOACs) versus vitamin K antagonists (VKAs) is a topic of ongoing debate. This study leverages a comprehensive national health database to evaluate the incidence of GIB, associated risk factors, and postbleeding management strategies among anticoagulated patients.

Materials and methods: Utilizing the Turkish Ministry of Health's e-Nabız system, we conducted a retrospective analysis of patients treated with DOACs and warfarin from January 2017 to July 2023. GIB events were identified using ICD codes, and comorbidities, prior medication use, interventions, and mortality rates were analyzed. Drug survival and patterns of changes following GIB were also evaluated.

Results: Among 102,545 patients with a GIB event during anticoagulant treatment, DOAC users were older with a higher prevalence of comorbidities, except for chronic obstructive lung disease, compared to VKA users. GIB-related mortality was 0.6% in the DOAC group and 0.4% in the VKA group at admission after the GIB (p < 0.01). In all drug groups, approximately half of the patients discontinued anticoagulation due to GIB after 3 months, the rate being highest with apixaban (61.9%). In patients who continued anticoagulation, the anticoagulant prior to GIB remained the most common agent in all groups, with rivaroxaban having the highest retention rate (40.7%).

Conclusion: This nationwide study indicates a higher frequency of GIB in DOAC users versus VKA users, with age and comorbidities potentially contributing to this trend. Mortality rates were comparable to the previous literature but warrant further investigation. The significant rate of discontinuation following GIB raises concerns about ongoing anticoagulation management. These findings underscore the need for cautious case management.