Objectives: We aimed to compare tuberculosis (TB) risk during biologics treatment between patients with RA who did (prophylaxis) and did not (non-prophylaxis) undergo chemoprophylaxis following pre-biologic latent TB screening in Korea of an intermediate TB burden.
Methods: Using the 2002-16 Korea National Health Insurance database, we conducted a cohort study examining TB risk, defined by International Classification of Diseases Tenth Revision codes plus anti-TB drugs, among RA patients initiating a biologic drug with and without chemoprophylaxis after screening triage for latent TB. To control baseline confounding, we used propensity score-based fine stratification (PSS) and weighting. Cox proportional hazards models estimated hazard ratios and 95% CIs comparing TB risk between the prophylaxis vs non-prophylaxis groups.
Results: The PSS-weighted study cohort (mean age 57.0 years; 81.3% female) included 2249 and 7225 RA patients in the prophylaxis and non-prophylaxis groups, respectively. During 2.42 years of biologics treatment, 118 patients developed TB with the incidence rate per 100 person-years of 0.33 in the prophylaxis and 0.63 in the non-prophylaxis groups. The PSS-weighted hazard ratio (95% CI) for TB associated with the prophylaxis was 0.52 (0.32, 0.86). During the follow-up time, the incidence rate of TB remained consistently low in the prophylaxis group but it was highest in the first year, then time-dependently declined in the non-prophylaxis group.
Conclusion: This population-based cohort study warns that the current screening-based preventive strategy generates a substantially higher TB risk after biologics initiation among screening-negative patients compared with screening-positive patients receiving chemoprophylaxis, when the background TB burden is not low.
Keywords: biologic; chemoprophylaxis; rheumatoid arthritis; screening; tuberculosis.
© The Author(s) 2020. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For permissions, please email: [email protected].