Background: Prednisolone predominantly causes hyperglycaemia between midday and midnight in hospitalized patients (1). Consequently, glargine-based basal-bolus insulin regimens may under-treat day-time hyperglycaemia and cause nocturnal hypoglycaemia. We investigated whether an isophane-based insulin regimen that delivers more insulin between midday and midnight is safer and more effective than a glargine-based insulin regimen.
Methods: We recruited 44 hospitalized patients prescribed ≥20 mg/day prednisolone acutely with one finger prick blood glucose level (BGL) ≥15 mmol/L or two BGLs ≥10 mmol/L in the prior 24 hours. Patients were randomised to receive insulin isophane before breakfast and insulin aspart before meals or insulin glargine before breakfast and insulin aspart before meals. The initial daily insulin dose in both groups was 0.5 U/kg body weight or 130% of the patients current daily insulin dose. 50% of the daily insulin dose was insulin aspart. Insulin regimens were adjusted daily using a standardized protocol. The primary endpoint was the percentage of time glucose was 4-10 mmol/L on day 1 of insulin treatment, assessed using a continuous glucose monitoring system.
Results: There were no significant differences in age, sex, BMI, diabetes, prior insulin treatment or glycosylated haemoglobin between the groups (Table). On day 1, there were no significant differences in prednisolone dose, time glucose was 4-10 mmol/L, hypoglycaemia, mean daily glucose or glucose during different time periods between the groups (Table). In patients treated for 3 days, prednisolone dose reduced (p=0.02) and insulin dose increased over time (p=0.02), but there were no significant change in the time glucose was 4-10 mmol/L (p=0.45) or mean glucose (p=0.15).
Conclusions: There was no difference in the efficacy or safety of isophane and glargine-based insulin regimens. Higher initial insulin doses and insulin dose adjustments are required to treat prednisolone-induced hyperglycaemia.