Objective: This study investigated whether the incidence of opportunistic infection differed in systemic lupus erythematosus patients who received different doses of corticosteroids. Methods: We included patients with diagnosed systemic lupus erythematosus from 1997 to 2010 using Taiwan national health insurance data. The index day for systemic lupus erythematosus patients was 3 months after the systemic lupus erythematosus diagnosis. A non-steroid cohort was matched 4:1 with the steroid cohort according to age, sex and index day. The end of the follow-up period was the day of opportunistic infection diagnosis, 1 year after the index day, or death. Results: The overall cumulative incidence of opportunistic infection was 136-fold higher in the steroid cohort than in the non-steroid cohort. The adjusted hazard ratio for developing mycobacterium infection in the steroid cohort was 11, and the adjusted hazard ratio for developing herpes zoster was 43.6 compared to the non-steroid cohort after adjusting for immunosuppressive agents and comorbidities. The adjusted hazard ratio value for opportunistic infection was 1.40 (95% confidence interval (CI) 0.78–2.51) for a daily prednisone-equivalent dose of 7.5–15 mg, 1.72 (95% CI 1.02–2.91) for 15–30 mg, 1.96 (95% CI 1.17–3.28) for 30–60 mg and 2.24 (95% CI 1.26–4.00) for over 60 mg compared with low-dose steroids (<7.5 mg). Conclusion: This study confirmed that the risk of opportunistic infection is higher in systemic lupus erythematosus patients treated with steroids in the first 3 months after diagnosis versus those not treated with steroids. Medium and high doses were associated with a higher risk of opportunistic infection compared with low doses. However, there was no controlling for disease activity, making it hard to know if increases in infection were due to disease itself or corticosteroids.
All Science Journal Classification (ASJC) codes