Patients undergoing glucocorticoid treatment are at risk of developing metabolic adverse effects, such as the metabolic syndrome. Preclinical studies have suggested that metformin can prevent these effects; a new paper has investigated the use of metformin in a randomized controlled trial. In the trial, 29 patients who did not have diabetes mellitus were randomly assigned to receive a glucocorticoid (prednisone, prednisolone or methylprednisolone) with a placebo or metformin (850 mg once daily for 1 week followed by 850 mg twice daily for 3 weeks). After 4 weeks, the median glucose levels increased in the placebo group but were similar to the baseline levels in the group that received metformin. The authors suggest that metformin could, therefore, be used to prevent metabolic adverse effects in patients taking glucocorticoids.