Background/Purpose: Autoimmune thyroid disease (AITD) is one of the most common autoimmune diseases, and occurs more often than expected in patients with Rheumatoid Arthritis (RA). The nature (shared susceptibility or risk factors, AITD triggering RA, or vice versa) and the pattern of risks in relation to RA onset remain unknown. We therefore aimed at assessing risks and relative risks of AITD as a function of time before and after, respectively, RA onset, overall and by age and sex.

Methods: In the Swedish Rheumatology Quality Register, we identified a large cohort of patients with incident RA (symptom duration at diagnosis < 1 year) 2006 through 2013 (n=8090). For each RA-patient 10 general population controls were individually matched. (n=80 856). Incident AITD was defined as a first ever prescription of thyroxin, based linkage to the Swedish Prescribed Drug register (excluding participants with prescriptions for iodine-containing drugs and thyroid cancer. We used a case-control approach to assess relative risks (odds ratios) for AITD before RA, and a cohort approach to assess relative risks (hazard ratios) of AITD after RA.

Results: By the time of RA diagnosis, 259 (3.2%) RA patients vs. 1613 (2.0%) of the general population controls had a history of AITD (OR=1.6, 95%CI 1.4-1.9). We noted increased risks for AITD up to five years before RA onset (OR=1.4, 95%CI 1.2-1.8), with the highest relative risks 3 months or less between the onset of AITD and the RA-diagnosis (OR=5.1, 95% CI 3.6-7.2). Separate analyses revealed higher relative risk among younger and seropositive patients. Following diagnosis of RA, AITD occurred among 121 (1.7%) RA-patients and 1384 (1.9%) controls, overall HR=0.9, 95% CI 0.7-1.1. The risk of AITD was increased during the first three months following RA diagnosis (HR= 2.1, 95% CI 1.2-3.7), but decreased over time into a decreased risk two or more years after RA diagnosis (HR=0.7, 95%CI 0.5-0.99).

Conclusion: This study confirms the increased risk of incident RA among patients with AITD, and extends this observation by demonstrating that the risk of incident AITD varies with time before/after RA, and that the increased risk before RA onset is replaced by a decrease in risk in patients with established RA. This temporal pattern of risk is compatible both with a critical link between AITD and RA onset, and with increased diagnostic intensity around the time-point of RA diagnosis. Whether the decrease in risk with increasing RA duration is a reciprocating decline after a transient increase in diagnostic intensity, or is a true protective effect e.g., via anti-rheumatic therapies, remains to be investigated.