Efficacy of Glimepiride for the Treatment of Diabetes Occurring During Glucocorticoid Therapy

Approximately 5–25% of patients receiving glucocorticoids exhibit overt diabetes (1). Glucocorticoids may precipitate diabetes in individuals with impaired insulin secretion by reducing insulin sensitivity (2). The relative importance of β-cell dysfunction and insulin resistance for the pathophysiology of the glucocorticoid-induced diabetes is not well defined. Therefore, there is no consensus treatment for glucocorticoid-induced diabetes. Glimepiride is a sulfonylurea that lowers blood glucose levels by stimulating insulin secretion from pancreatic β-cells and secondarily by increasing glucose uptake in peripheral tissues (3). Such action mechanisms might be suitable for the treatment of glucocorticoid-induced diabetes. Here we examined the effects of glimepiride on patients with newly diagnosed diabetes during glucocorticoid therapy.

Three Japanese female patients who had been taking oral glucocorticoids were newly diagnosed with diabetes. Patient 1 (aged 68 years) had systemic lupus erythematodes, patient 2 (aged 65 years) had Behcet’s disease, and patient 3 (aged 48 years) had angiolymphoid hyperplasia with eosinophilia. They had been initially given 20–40 mg/day prednisolone. The dosage of prednisolone was tapered and maintained at 5–10 mg/day. The status of these diseases was well controlled with the glucocorticoid treatment. At 1–2 years after starting the glucocorticoid therapy, they showed overt diabetes, with mean fasting blood glucose 12.6 ± 0.7 mmol/l and HbA_1c 9.5 ± 1.5% (means ± SE). Their index for pancreatic β-cell function (HOMA-%β), as determined by the correct homeostasis model assessment evaluation (4), was 27 ± 8%, significantly (P < 0.001) lower than that (72 ± 4% [range 44–111%]) in 24 healthy Japanese control subjects (mean age 47 ± 2 years) who had normal glucose tolerance by 75-g oral glucose tolerance test. The index for insulin sensitivity (HOMA-%S) (4) was 56 ± 15% in the patients, significantly (P < 0.01) lower than in healthy control subjects (144 ± 10% [81–273%]). We administered glimepiride to these patients (1 mg/day for patient 1 and 3 mg/day for patients 2 and 3). The dosage of prednisolone was unchanged throughout the observation period (24 weeks). Fasting blood glucose declined 4 weeks after the glimepiride administration and was kept below 7 mmol/l until 24 weeks. HbA_1c significantly decreased 4 weeks after the treatment, decreasing to 6.7 ± 0.6% after 8 weeks and maintaining that level until 24 weeks. HOMA-%β and HOMA-%S increased to the control levels (76 ± 9% and 108 ± 54%, respectively) 8 weeks after the treatment, remaining within the control ranges at 24 weeks (60 ± 3% and 99 ± 55%).

In three patients with glucocorticoid-induced diabetes, HOMA-%β and HOMA-%S were lower than in healthy control subjects. After the treatment with glimepiride, HOMA-%β and HOMA-%S increased to control ranges, in association with remarkable improvement of glycemic controls persisting until 24 weeks. This is suggested to be attributable to the dual effects of glimepiride on β-cell function and insulin sensitivity. It has been shown that a thiazolidinedione has a potential for the glucocorticoid-induced diabetes (5). From our study, glimepiride is also a strong candidate for the treatment of the glucocorticoid-induced diabetes.