Abstract
Background: Glucocorticoid-induced hyperglycemia (GCIH) is a common adverse effect of glucocorticoid (GC) therapy. Although evidence on oral antidiabetic medications (OADMs) for GCIH is emerging, direct comparisons with insulin therapy remain limited. This study aimed to compare the efficacy and safety of OADMs and sliding scale insulin (SSI) in patients with autoimmune diseases who developed GCIH. Methods: We retrospectively analyzed 97 patients who developed GCIH during GC therapy equivalent to ≥20 mg/day of prednisolone. Patients were classified into SSI-only (n = 41), OADM (n = 31), and basal-bolus/basal or bolus insulin (BBI/BI) (n = 25) groups. The primary endpoint was mean preprandial blood glucose (BG), adjusted for patient characteristics. Secondary outcomes included hospital stay, hypoglycemia, insulin use, and glycated hemoglobin. Results: In univariate analysis, the mean preprandial BG levels during the treatment period were significantly associated with the mean initial preprandial BG levels. Adjusted mean preprandial BG during treatment did not differ significantly between the OADM and SSI-only groups, whereas the BBI/BI group had higher pre-breakfast BG (p = 0.016). Among first-time GC users, those in the OADM group using cyclophosphamide had significantly lower fasting BG than non-users (p = 0.011). Conclusions: In patients with autoimmune diseases receiving ≥20 mg/day GC, OADM provided glycemic control comparable to SSI with similar hypoglycemia risk. Early preprandial BG levels during the first 3 days of GC therapy may help predict glycemic outcomes. Prospective studies with standardized regimens are needed to optimize GCIH management.