Effects of low or medium-dose of prednisone on insulin resistance in patients with systemic lupus erythematosus.

OBJECTIVES To determine the effect of low (< or =7.5 mg/day; LD-PRD group) or medium (>7.5 mg/day; MD-PRD group) doses of prednisone over the past 4 months on insulin levels and insulin resistance (IR) in SLE patients. METHODS SLE patients were categorised in prednisone non-users (No PRD) (n=41), LD-PRD (n=71) and MD-PRD (n=16) users. We compared insulin levels, presence of increased IR using homeostasis model assessment (HOMA index), metabolic syndrome (MetS), and other clinical, metabolic and inflammatory parameters in the 3 groups. A Spearman's rho test was used to identify independent associations between daily prednisone dose, HOMA index and insulin levels and other parameters, after adjusting for confounders. RESULTS No differences in increased IR, HOMA index and insulin between No PRD and LD-PRD were found. In contrast, the MD-PRD group was younger (p=0.001) and had higher insulin (p=0.015), higher HOMA index (p=0.019) and increased IR (OR 5.8, 95% CI (1.7-20), p=0.007) in comparison with the LD-PRD group. The HOMA index strongly correlated with body mass index (BMI) (rs=0.460, p<0.001) but not with clinical activity or inflammatory state after adjusting for confounders. Prednisone dose correlated with the HOMA index and insulin but not with inflammatory parameters (erythrocyte sedimentation rate p=0.075) after adjusting for confounder. CONCLUSIONS Daily medium-dose prednisone use (>7.5 mg/d) but not low-dose (< or =7.5 mg/d) use increased insulin levels and IR in SLE, which may contribute to increased CV risk experienced by these patients.