Risk of Ankylosing Spondylitis in Patients With Endometriosis: A Population-Based Retrospective Cohort Study.

Objectives: Previous research has shown a possible relationship between endometriosis and autoimmune diseases. However, the relationship between endometriosis and ankylosing spondylitis (AS) is lacking. Therefore, we intended to find possible associations between endometriosis and AS using ICD-9 coding data in a population-based retrospective cohort study in Taiwan. Method: Data for this retrospective cohort study were collected from the Taiwan National Health Insurance Research Database (NHIRD) between 2000-2012. We collected 13,145 patients with endometriosis and a 78,870 non-endometriosis comparison cohort. Diagnoses of endometriosis and AS were defined by the International Classification of Diseases-9 (ICD-9-CM) code for at least 3 outpatients or 1 hospitalization. Propensity score matching by comorbidities, corticosteroids, and non-steroidal anti-inflammatory drugs (NSAIDs) usage were done for baseline comparability. Cox proportional hazard models were used to evaluate crude and adjusted hazard ratios. Results: The cumulative incidence of AS was higher in patients with endometriosis compared to the non-endometriosis comparison cohort (log-rank test, p = 0.015). The adjusted hazard ratio (aHR) of incidental AS in patients with endometriosis was 1.61 (95% CI = 1.11 to 2.35) in comparison to the non-endometriosis comparison cohort. An increased risk of AS was also observed in subjects with major depressive disorder (aHR = 5.05, 95% CI = 1.85 to 13.78). Stratified analyses of age subgroups showed consistent results. NSAID users had a lower risk of AS than NSAID non-users (aHR 4.57 vs 1.35, p for interaction = 0.031). Conclusions: In this retrospective population-based cohort study, we found a higher risk of AS in patients with endometriosis. We suggest that clinicians should pay attention to the occurrence of AS in patients with endometriosis.