Glucocorticoids are anti-inflammatory drugs used in combination with inhaled bronchodilators, such as β2-agonists and antimuscarinics, for the treatment of stable chronic obstructive pulmonary disease (COPD), to improve respiratory symptoms, such as exertional dyspnoea, and to decrease the risk of future COPD exacerbations. However, it remains controversial whether their regular long-term use increases the risk of developing diabetes mellitus. The objective of this narrative review is therefore to analyse all the randomized controlled trials performed in patients with stable COPD to identify the risk of new onset diabetes mellitus during a long-term (at least 52 weeks) regular treatment with inhaled glucocorticoids alone compared to placebo. From a literature search on PubMed, 19 studies fulfilling these criteria have been identified. The inhaled glucocorticoids administered were: fluticasone propionate (7 studies), budesonide (6 studies), mometasone furoate (3 studies), beclomethasone dipropionate (1 study), triamcinolone acetonide (1 study), and fluticasone furoate (1 study) respectively. Only 3 out of the 19 trials identified in our narrative review reported data on diabetes mellitus, and in these the incidence of diabetes mellitus was not significantly different in both treatment arms (inhaled glucocorticoids and placebo), regardless of the type of glucocorticoid used.