Postdoc Department of Clinical Epidemiology, Aarhus University Hospital and Aarhus University, Denmark
Background: Both migraine and non-steroidal anti-inflammatory drugs (NSAIDs) have been associated with increased risk of ischemic stroke (IS) and myocardial infarction (MI). NSAIDs are recommended as first-line treatment for acute migraine, but no study has examined the cardiovascular safety of using NSAIDs for migraine. Moreover, it is unknown whether the increased arterial thromboembolic risk in patients with migraine can be explained partly or fully by NSAID use in these patients.
Objectives: To examine (1) the risk of IS and MI associated with NSAID use in patients with migraine and (2) whether the risk of IS and MI associated with migraine is modified by NSAID use.
Methods: We identified all patients with an incident hospital migraine diagnosis ≥18 years of age during 1995–2020 without a history of IS or MI (n=70,720). We matched these patients with five individuals from the general Danish population without migraine or a history of IS or MI on birth year and sex (n=353,600). We followed the individuals from the migraine/matching date until the first of an outcome, death, emigration, or December 31st, 2020. NSAID use was determined in a time-varying manner. We calculated (1) hazard ratios (HRs) with 95% confidence intervals of the association between NSAID use and the outcomes and (2) incidence rate ratios (IRRs) with 95% confidence intervals of the association between migraine and the outcomes stratified by NSAID use.
Results: Patients with migraine were exposed to NSAIDs during 12% of the follow-up compared with 8.3% among comparison cohort members. The HRs associating NSAID use with IS were 1.19 (1.10; 1.29) in patients with migraine and 1.37 (1.31; 1.42) in comparison cohort members. The corresponding HRs for MI were HR=1.18 (1.08; 1.28) in patients with migraine and 1.38 (1.32; 1.44) in comparison cohort members. The IRRs of the association between migraine and IS were 1.82 (1.74; 1.90) overall, 1.35 (1.18; 1.54) among NSAID users, and 1.84 (1.76; 1.93) among NSAID non-users. The corresponding IRRs for MI were 1.40 (1.31; 1.50) overall, 1.02 (0.83; 1.24) among NSAID users, and 1.42 (1.32; 1.52) among NSAID non-users.
Conclusions: As in the general population, NSAID use was in patients with migraine also associated with increased risk of arterial thromboembolic events. NSAID use, however, could not explain the arterial thromboembolic risks associated with migraine itself.
