The association between Candida infection and ankylosing spondylitis: a population-based matched cohort study.

AIMS To explore whether newly diagnosed Candida infection with risk of developing ankylosing spondylitis (AS). Methods and materials: We investigated 61,550 patients with newly diagnosed Candida infection between 1997 and 2013 from the Taiwan National Health Insurance Research Datasets to conduct a population-based matched-cohort study. Controls were 61,550 subjects without Candida infection and propensity score matched with the Candida exposure cohort. The follow-up period was defined as month from the initial diagnosis of Candida infection (or nested index date for controls) to the date of AS, or 31 December 2013. We used Cox proportional hazards models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for the occurrence of AS. RESULTS The incidence rates of AS in the Candida group and comparison groups were 4.58 and 3.88 per 100,000-person months, respectively. The adjusted HR (95% CI) of AS for the Candida group was 1.19 (0.99-1.44) compared to the control group after adjustment for age, gender, and all covariates. (95% CI =1.77-2.27). However, an aHR of 1.77-fold (95% CI: 1.26-2.53) significant increase in the risk of developing AS was observed after 6 years of follow-up, when exposure to Candida was at baseline. The effect of Candida infection was significantly time varying (P value for interaction between follow up period and Candida infection is 0.018). CONCLUSIONS A risk of AS was found after Candida infection, and year of follow up acts as an effect modifier between the Candida infection and risk of AS. Key message: What is already known on this subject? Links between spondyloarthritis and fungal infections have been found in animal studies before. What does this study add? Our study demonstrated that Candida infection is an independent risk factor for developing ankylosing spondylitis in terms of gender, age, and relevant variables and comorbidities. A risk of ankylosing spondylitis was found after Candida infection, and year of follow up acts as an effect modifier between the Candida infection and risk of AS. Clinicians should be aware of possible Candida infection in managing patients with ankylosing spondylitis. Implications: Clinicians must pay greater attention to patients with newly diagnosed Candida infection. Specifically, they should conduct tests for ankylosing spondylitis. Further research is needed to examine if and how treatment of Candida infection alleviates symptoms of AS.