Heightened migraine risk in patients with rheumatoid arthritis: A national retrospective cohort study

ObjectiveThis study aimed to evaluate the association between rheumatoid arthritis (RA) and subsequent migraine risk using the Korean National Health Insurance Service database.BackgroundMigraine may be related to immune dysfunction and previous studies have suggested an association with chronic inflammatory rheumatic diseases; however, the relationship between RA and migraine remains unclear.MethodsThis was a population‐based, nationwide, retrospective, longitudinal cohort study. Participants were enrolled from 2010 to 2017 and followed up until 2019. A total of 42,674 patients who had undergone a health checkup within 2 years prior to the initial diagnosis of RA were included in the study, after applying the exclusion criteria (previous migraine, other rheumatic disease, missing variables of interest). A non‐RA control was obtained by age and sex‐matching (1:5). Finally, 42,644 patients with RA were enrolled, with 213,370 individuals without RA included as controls. Among the patients with RA, 29,744 had seropositive RA (SPRA), and 12,900 had seronegative RA (SNRA). SPRA was defined by the International Classification of Diseases 10th revision (ICD‐10) code M05, prescription of disease‐modifying anti‐rheumatic drugs (DMARDs), and enrollment in a special copayment reduction program. SNRA was defined by the ICD‐10 code M06 and prescription of any DMARD. The primary endpoint was the occurrence of migraine incidents, defined using the ICD‐10 code of migraine (G43).ResultsA total of 22,294 migraine cases (17,912/213,370 [8.3%] in controls and 4382/42,674 [10.2%] in RA) were reported during a mean follow‐up of 4.4 years after a 1‐year lag period. Patients with RA had a 1.2‐fold higher risk of migraine compared with controls (adjusted hazard ratio [aHR] 1.21, 95% confidence interval [CI] 1.17–1.26). Increased risk of migraine was found in both patients with SNRA and SPRA compared with controls (aHR 1.20, CI 1.15–1.24 in SPRA; aHR 1.26, CI 1.19–1.34 in SNRA). Compared to patients with SNRA, those with SPRA did not demonstrate a heightened risk (aHR 0.94, CI 0.88–1.01). A significant interaction was confirmed between covariates (male, current smoker, those with diabetes mellitus, and dyslipidemia) and the risk of migraine (p for interaction of <0.05).ConclusionRA was linked to a higher migraine risk, regardless of seropositivity.