Abstract
BACKGROUND: Biological disease-modifying antirheumatic drugs (bDMARDs) and targeted synthetic (ts) DMARDs are important in rheumatoid arthritis (RA) treatment. The risk of hospitalized infection associated with bDMARDs/tsDMARDs in RA patients is unclear. METHODS: We retrospectively analyzed the cases of the 275 RA patients with 449 treatment episodes who were administered a bDMARD/tsDMARD at Nagasaki University Hospital in July 2003-January 2015. We determined the incidence and risk factors of infection requiring hospitalization in the patients during a 3-year observation period. RESULTS: Thirty-five (12.7%) of the patients experienced a hospitalized infection. The hospitalized infection risk did not differ significantly among several bDMARDs/tsDMARDs. A multivariate analysis revealed that the comorbidities of chronic lung disease (adjusted HR 5.342, 95% CI 2.409-12.42, p < 0.0001) and the initiation of bDMARDs/tsDMARDs before 2010 (adjusted HR 4.266, 95% CI 1.827-10.60, p = 0.0007) are significant independent risk factors for hospitalized infection. Compared to the before-2010 group, the group of patients whose treatment initiated in 2010 or later showed higher patient ages at the initiation of bDMARD/tsDMARD treatment and a higher rate of the use of prophylaxis with an antituberculosis agent, whereas the disease activities and number of the patients who received >5 mg of prednisolone were lower in the after-2010 group. CONCLUSIONS: This is the first report that the frequency of hospitalized infection significantly decreased when the patients were treated with a bDMARD or tsDMARD after 2010. Our results indicate that the updated announcement of diagnosis and treatment criteria might contribute to a reduced risk of hospitalized infection and a better understanding of the use of bDMARDs/tsDMARDs by rheumatologists.