Background: Selective serotonin reuptake inhibitors (SSRIs) are the first-line treatment for depression but there is a concern about the risk of upper gastrointestinal bleeding (UGIB). Past studies, however, are largely confounded by the presence of Helicobacter pylori (HP).
Aim: To evaluate the UGIB risk of SSRI users after treatment for HP.
Methods: This was a propensity score (PS) matched cohort study with patients who used SSRI after receiving HP eradication therapy from the Hong Kong territory-wide healthcare database. The primary outcome was hospitalisation for nonvariceal UGIB. PS matching analysis with a ratio of 1:2 plus Cox regression model was used to compute the hazards ratios (HR) and 95% CI of UGIB risk.
Results: In this study, 3358 SSRI users and 57 906 non-users were included. The median follow-up duration was 7.74 (interquartile range 5.32-10.42) years. The crude incidence of hospitalisation for UGIB was 3.98 (95% CI 3.80-4.16) per 1000 person-years. In the PS matching analysis of 3358 SSRI users with 6716 non-users, SSRI was associated with a higher risk of UGIB compared to non-users (HR 1.95, 95% CI 1.41-2.70). This result was consistent in sensitivity analysis with 1:1 PS matching (HR 2.13, 95% CI 1.50-3.02) and multivariable analysis with 1-month intervals (HR 1.81, 95% CI 1.34-2.45) or 3-month intervals (HR 1.61, 95% CI 1.20-2.17). After stratifying by age, the increased risk of SSRI was only significant among patients >50 years.
Conclusion: SSRI users have a higher risk of hospitalisation for nonvariceal UGIB after treatment for HP, particularly among older patients.
© 2019 John Wiley & Sons Ltd.