Hepatitis B virus-related mortality in rheumatoid arthritis patients undergoing long-term low-dose glucocorticoid treatment: A population-based study

Abstract

Glucocorticoids (GC) are commonly used in rheumatoid arthritis (RA) patients which bears a risk of hepatitis B virus (HBV) reactivation. The purpose of this study was to investigate the risk of HBV-related mortality under long-term low-dose GCs in Taiwanese RA patients. We retrospectively analyzed 45,423 RA patients using National Health Insurance Research Database from January 1999 to December 2011. Of them, 2204 patients had the diagnosis of HBV and were classified into four groups according to GCs regimens. Outcome comparison by Cox model analysis for liver-related mortality was performed. In this cohort, 90.5% were older than 40. One hundred and five patients had been treated with short-term large-dose GCs (Group A); 862 patients received GCs ≥20 mg/day for ≥3 days or a variable dose but did not meet Group C criteria (Group B); 689 patients were continuously treated with low-dose (<20mg/day) GCs for ≥3 months for at least one session (Group C); and 548 patients had never been exposed to GCs (Group D). Two hundred and sixty-one patients had been exposed to antiviral agents, which was significantly higher in Group C. Fifty-eight patients (2.63%) died of acute hepatic failure, while no statistically significant difference between each groups (p=0.074). Groups C and D comparison by two-sample test showed that long-term low-dose GC treatment was not associated with liver-related death after adjusting for malignancy. Long-term low-dose GC treatment was not associated with liver-related mortality in RA with concomitant HBV patients probably due to commonly applied antiviral therapy by rheumatologists.