Abstract
Background and aim
Direct oral anticoagulants (DOACs) became a widespread alternative in anticoagulant therapy. Nevertheless, concerns are raised about their safety, with increased gastrointestinal bleeding rates being described. There are scarce studies regarding DOACs effect on small-bowel capsule endoscopy (SBCE) findings. We aimed to assess if the detection of lesions with high bleeding potential on SBCE was significantly different in patients treated with DOACs when compared to non-anticoagulated patients and to patients anticoagulated with other agents.
Methods
Cohort study including consecutive patients who underwent SBCE for suspected mid-gastrointestinal bleeding (MGIB) in 2019 and 2020.
Results
From 148 patients, 38 (25.7%) were anticoagulated, of which 26 (68.4%) with DOACs. P2 lesions were detected in 36.5% (n = 54) of the patients. These lesions were more frequently detected in patients under DOACs treatment when compared to non-anticoagulated patients (69.2% vs. 29.1%; p=.001), and also when compared to patients treated with other anticoagulants (69.2% vs. 33.3%; p=.037). No differences in P2 lesions detection were observed between patients treated with other anticoagulants and non-anticoagulated patients (33.3% vs. 29.1%; p=.747). In multivariate analysis, DOACs usage was significantly associated with higher detection rates of P2 lesions on SBCE, when adjusted for classical risk factors for MGIB (OR: 3.38; 95%CI = 1.23–9.26; p=.018).
Conclusions
Despite their undeniable cardiovascular benefits and easy applicability, DOACs should still be considered with caution. These drugs were significantly associated with higher risk of potentially bleeding lesions on SBCE when compared to other anticoagulants and represent an independent risk factor for MGIB when adjusted for other variables.
Authors contributions
All authors contributed to and agreed on the final content of the manuscript. V. Macedo Silva designed the study, collected and carried out data analysis and drafted the manuscript. M. Freitas and C. Arieira collected and carried out data analysis. S. Xavier, P. Boal Carvalho, B. Rosa and MJ. Moreira critically revised the manuscript. J. Cotter critically revised and approved the final version of the manuscript.
Disclosure statement
The authors report no conflict of interest.