Bleeding risk associated with non-steroidal anti-inflammatory drug use in patients with venous thromboembolism: a nationwide cohort study

Background: The bleeding risk of using non-steroidal anti-inflammatory drugs (NSAIDs) in patients treated with oral anticoagulants for venous thromboembolism remains unclear.

Objectives: To examine the bleeding risk associated with use of NSAIDs in patients treated with oral anticoagulants for venous thromboembolism.

Methods: Using Danish health registries, we conducted a nationwide population-based cohort study of all adult patients initiating an oral anticoagulant (a direct oral anticoagulant or warfarin) after a first-time diagnosis of venous thromboembolism between 1 January 2012 and 30 June 2022. We excluded patients with a history of bleeding. The primary outcome was a bleeding composite defined from hospital-diagnosed gastrointestinal, intracranial, thoracic, respiratory tract, or urinary tract bleeding, or anaemia caused by bleeding. Secondary outcomes included site-specific bleeding. We calculated crude event rates and used multivariate Cox regression to compute adjusted hazard ratios for the association between NSAID use (modelled in time-varying manner) and the outcomes.

Results: Among 49,076 patients (median age, 69 years; 48% female), 11% redeemed at least 1 prescription for an NSAID. Event rates for any bleeding per 100 person-years were 3.6 (95% confidence interval [95% CI] 3.4–3.8) during periods without NSAID use and 6.5 (95% CI 5.2–8.1) during periods with NSAID use. Compared with non-use, the adjusted hazard ratios for any bleeding associated with NSAID use were 2.11 (95% CI, 1.68–2.66) overall, 1.78 (95% CI, 1.34–2.36) for ibuprofen, 3.47 (95% CI, 1.91–6.29) for diclofenac, and 4.19 (95% CI, 2.17–8.07) for naproxen. Compared with non-use, the adjusted hazard ratios associated with NSAID use were 2.24 (95% CI, 1.60–3.14) for gastrointestinal bleeding, 3.30 (95% CI, 1.73–6.31) for intracranial bleeding, 1.41 (95% CI, 0.69–2.87) for thoracic and respiratory tract bleeding, 1.56 (95% CI, 0.96–2.54) for urinary tract bleeding, and 2.98 (95% CI, 1.44–6.17) for anaemia caused by bleeding. Results were consistent in subgroups of patients using a direct oral anticoagulant or warfarin and with deep vein thrombosis or pulmonary embolism.

Conclusions: Patients treated with oral anticoagulants for venous thromboembolism had a more than twofold increased bleeding rate when using NSAIDs. The increased risk of bleeding was not restricted to the gastrointestinal tract but applied to all organ systems.