Prednisone and long-term damage in systemic lupus erythematosus: Which is the threshold dose? A pilot study.

OBJECTIVES

We aimed to study the daily dose of glucocorticoids in the first month associated with damage after 5 years of systemic lupus erythematosus (SLE) diagnosis; and to assess the daily dose of prednisone during the first year, over which damage after 5 years could be predicted.

METHODOLOGY

A retrospective cohort study of SLE patients from the diagnosis and up to 5 years of follow-up was performed. Cumulative prednisone doses in the first month (T1m), at 1 year (T12m), and at 5 years (T5y) were calculated. Damage was estimated using the SLICC index and activity by the SLEDAI-2K. Damage at T5y was categorized as "present" or "absent". To determine the cutoff point of prednisone dose for the first month and for the first year of treatment, related to damage at 5 years, a ROC curve was done. A logistic regression was performed to control damage at 5 years by the activity levels, anti-DNA titers, and the corticosteroid burden between 1 and 5 years.

RESULTS

Forty-eight patients were included with a mean age of 42.4 (12.6) years; 46 (95.8%) were female. The cumulative dose in T1m associated with greater sensitivity and specificity in the ROC curve (1.54; AUC 0.793; 95% CI: 0.66-0.92) to predict glucocorticoid-related damage at 5 years of follow-up was 980mg, which leads to a daily dose for the first month of 32.6 mg/day. The daily dose during the first year associated with damage at 5 years, allowed to calculate an optimal cutoff point of 7.38 mg/day, with a sensitivity of 92.9% (AUC 0.695, 95% CI: 0.52-0.86) to predict damage at 5 years. In the logistic regression, this estimate was maintained, controlling for SLEDAI-2K and for positivity or not of anti-DNA antibodies and for the cumulative dose of prednisone between T5y and T12m.

CONCLUSION

This pilot study allows to estimate a threshold dose of 32.6 mg/day during the first month and 7.38 mg/day during the first year, over which the risk of permanent damage is high at 5 years, regardless of the activity levels and the glucocorticoid doses considered between the first and the fifth year of follow-up.