There were no differences in the efficacy or safety of the isophane and glargine-based insulin regimens. We recommend an initial daily insulin dose of 0.5 units/kg bodyweight if not on insulin, a greater than 30% increase in pre-prednisolone insulin dose and larger insulin dose adjustments in patients with prednisolone-induced hyperglycaemia.

Glucocorticoids are frequently prescribed to patients with a wide range of inflammatory and autoimmune diseases. The semi-synthetic glucocorticoid prednisolone is most commonly prescribed and in two main patterns. Prednisolone is prescribed short-term at medium-high doses to treat an acute inflammatory illness or long-term at lower doses to attenuate chronic inflammatory disease progression. In hospitalized patients with acute prednisolone-induced hyperglycaemia there is a distinct circadian pattern of glucose elevation, which occurs predominantly in the afternoon and evening. As a morning dose of isophane insulin has a pharmacokinetic pattern that matches this pattern of glucose elevation, treatment comprising a basal dose of morning isophane insulin in combination with short-acting insulin boluses is generally recommended. However, evidence is lacking that isophane-based basal bolus insulin is more efficacious than other insulin regimens. In outpatients, low dose prednisolone causes a small increase in post glucose-load glucose concentration but no change in overall glycaemic control as measured by glycosylated haemoglobin. If treatment is indicated, metformin has been shown to be effective and may attenuate other adverse effects of long-term prednisolone therapy. Further studies are necessary in order to identify factors underlying the variability in response to insulin therapy and clinical benefits of treatment in hospitalized patients with prednisolone-induced hyperglycaemia. In outpatients prescribed low dose prednisolone, the cardiovascular risk associated with postprandial hyperglycaemia and efficacy of hypoglycaemic therapies should be evaluated in future randomized clinical trials.