No increased risk of herpes zoster in TNF inhibitor and non-TNF inhibitor users with rheumatoid arthritis: epidemiological study using the Japanese health insurance database

Objective It is controversial whether the use of biological disease-modifying antirheumatic drugs (DMARDs) increases the risk of herpes zoster (HZ). We aimed to evaluate the risks of HZ in tumor necrosis factor inhibitor (TNFI) and non-TNFI users with rheumatoid arthritis (RA) over 3 years in Japan. Method Using the Japanese health insurance database, we assigned patients with at least one RA diagnostic code and one prescription for any DMARDs (RA cases) recorded between January 2005 and December 2013 to the RA group. We randomly selected five age-, sex-, calendar year- and observation length-matched non-RA cases for each RA case (non-RA group), and assessed associations between RA and HZ. To evaluate the risks of HZ in TNFI and non-TNFI users, we conducted a nested case-control study (NCC) in the RA group. Results The RA group (n = 6712) had a significantly higher crude incidence rate of HZ than the non-RA group (n = 33 560) (14.2 vs. 8.3/1000 patient-years), and the adjusted odds ratio (95% confidence interval) of the RA versus non-RA groups was 1.43 (1.17-1.75). The NCC demonstrated that use of TNFI, non-TNFI, methotrexate, or immunosuppressive DMARDs did not increase the risks of HZ. Use of corticosteroid ≥ 5 mg/day conveyed a significant risk of HZ in patients with RA. Conclusions Rheumatoid arthritis was significantly associated with the development of HZ, and use of corticosteroids ≥ 5 mg/day was identified as a significant risk factor, whereas either TNFI or non-TNFI use were not.