Premium
Gastrointestinal bleeding risk with rivaroxaban vs aspirin in atrial fibrillation: A multinational study
Author(s) -
Fanning Laura,
Wong Ian C. K.,
Li Xue,
Chan Esther W.,
Mongkhon Pajaree,
Man Kenneth K. C.,
Wei Li,
Leung Wai K.,
Darzins Peteris,
Bell J. Simon,
Ilomaki Jenni,
Lau Wallis C. Y.
Publication year - 2020
Publication title -
pharmacoepidemiology and drug safety
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.023
H-Index - 96
eISSN - 1099-1557
pISSN - 1053-8569
DOI - 10.1002/pds.5130
Subject(s) - rivaroxaban , medicine , aspirin , atrial fibrillation , gastrointestinal bleeding , propensity score matching , cohort study , cohort , warfarin
Purpose Comparative gastrointestinal bleeding (GIB) risk between rivaroxaban and low‐dose aspirin is unknown in patients with atrial fibrillation (AF). This study investigated GIB risk with rivaroxaban vs aspirin among two separate AF cohorts in Hong Kong and the United Kingdom, using a common protocol approach. Methods This was a population‐based cohort study using separate data from the Clinical Data Analysis and Reporting System (CDARS) of the Hong Kong Hospital Authority (2010‐2018) and The Health Improvement Network (THIN) database in the United Kingdom (2011‐2017). Patients with AF newly prescribed aspirin or rivaroxaban were included. Cox proportional hazards regression was used to compare GIB risks for rivaroxaban vs aspirin, accounting for confounders using propensity score fine stratification approach. Results In CDARS, 29 213 patients were included; n = 1052 (rivaroxaban), n = 28 161 (aspirin). Crude GIB event rates per 100 patient‐years in CDARS were 3.0 (aspirin) and 2.6 (rivaroxaban). No difference in GIB risk was observed between rivaroxaban and aspirin overall (HR = 1.04, 95%CI = 0.76‐1.42), and in dose‐stratified analyses (HR = 1.21, 95%CI = 0.84‐1.74 [20 mg/day]; HR = 0.80, 95%CI = 0.44‐1.45 [≤15 mg/day]). In THIN, 11 549 patients were included, n = 3496 (rivaroxaban) and n = 8053 (aspirin). Crude GIB event rates were 1.3 (aspirin) and 2.4 (rivaroxaban) per 100 patient‐years. No difference in GIB risk was observed between rivaroxaban and aspirin overall (HR = 1.40, 95%CI = 1.00‐1.98) and low‐dose rivaroxaban (≤15 mg/day) (HR = 1.00, 95%CI = 0.56‐1.30), but increased GIB risk was observed for rivaroxaban 20 mg/day vs aspirin (HR = 1.57, 95%CI = 1.08‐2.29). Conclusion In patients with AF, GIB risk was comparable between aspirin and rivaroxaban ≤15 mg/day. GIB risk for rivaroxaban 20 mg/day vs aspirin remains uncertain and warrants further investigation.