Incidences, temporal trends and risks of hospitalisation for gastrointestinal bleeding in new or chronic low-dose aspirin users after treatment for <i>Helicobacter pylori</i>: a territory-wide cohort study.

Abstract

The risk of GI bleeding (GIB) in aspirin users after <i>Helicobacter pylori</i> (HP) eradication remains poorly defined. We characterised the incidences and temporal trends of hospitalisations for all GIB in aspirin users after HP eradication therapy. Based on a territory-wide health database, we identified all patients who had received the first course of clarithromycin-based triple therapy between 2003 and 2012. Patients were divided into three cohorts according to aspirin use: new users (commenced after HP eradication), chronic users (commenced before and resumed after HP eradication) and non-users. The primary outcome was to determine the risk of hospitalisation for GIB. We included 6985 new aspirin users, 5545 chronic users and 48 908 non-users. The age-adjusted and sex-adjusted incidence of hospitalisation for all GIB in new, chronic and non-users was 10.4, 7.2 and 4.6 per 1000 person-years, respectively. Upper and lower GIB accounted for 34.7% and 45.3% of all bleeding, respectively. Compared with chronic users, new users had a higher risk of GIB (HR with propensity score matching: 1.89; 95% CI 1.29 to 2.70). Landmark analysis showed that the increased risk in new aspirin users was only observed in the first 6 months for all GIB (HR 2.10, 95% CI 1.41 to 3.13) and upper GIB (HR 2.52, 95% CI 1.38 to 4.60), but not for lower GIB. New aspirin users had a higher risk of GIB than chronic aspirin users, particularly during the initial 6 months. Lower GIB is more frequent than upper GIB in aspirin users who had HP eradicated.