Presented at the 2012 AAPM Annual Meeting « Back

The Risk of Bleeding with Duloxetine Treatment in Patients Who Use Nonsteroidal Anti-Inflammatory Drugs

Elijah P. Frakes, PhD, efrakes@lilly.com¹, Qi Zhang, PhD², Kristin Bullok, PhD¹, Michael J. Adams, BSN RN¹, Lori M. Miller, MS¹, Mark E. Bangs, MD¹, David G. Perahia, MD³, (1) Eli Lilly and Company, Indianapolis, Indiana, (2) Statistics, Indianapolis, Indiana, (3) Neuroscience, Windlesham, Surrey, United Kingdom

Introduction/Statement of Problem: Bleeding events are associated with selective serotonin reuptake inhibitor (SSRI) and serotonin/norepinephrine reuptake inhibitor (SNRI) use. Concomitant nonsteroidal anti-inflammatory drug (NSAID) use may add to this risk. This analysis focused on duloxetine safety surrounding bleeding events in patients with concomitant NSAID use or non-use. Materials and Methods: Placebo-controlled trial data were searched for bleeding-related events in patients, stratified by NSAID use. Differences between NSAID subgroups were analyzed using a logistic regression model including therapy, NSAID use, and therapy-by-NSAID subgroup interaction. Tests of hypotheses were based on a two-sided significance of 0.05, excepting 0.1 for therapy-by-NSAID subgroup interaction. Results: Treatment-by-NSAID subgroup interaction was significant for occurrence of at least one bleeding related event (p = .029), with significantly greater difference in rates between duloxetine/placebo (1.51% vs. 0.80%, respectively) in NSAID non-users compared with duloxetine/placebo difference (2.35% vs. 2.13%, respectively) in NSAID users. NSAID users experienced significantly greater frequency of bleeding events as compared with NSAID non-users (p<.001). Among NSAID users, no significant difference (p = .598) in bleeding event incidence was observed between placebo and duloxetine. No significant treatment-by-NSAID subgroup interaction was observed for gastrointestinal (GI) bleeding-related events combined (p = .742), with no significant differences between duloxetine and placebo in NSAID users and non-users. Conclusions: Concomitant NSAID use was associated with increased bleeding events in both treatment groups. For duloxetine or placebo NSAID users, no significant difference in bleeding event incidence was observed. No difference was observed for GI-related bleeding event incidence overall with regards to use of NSAIDs or duloxetine versus placebo.

Funding: All are employees of Eli Lilly and Company and/or one of its subsidiaries.