OBJECTIVES: Glucocorticoid treatment is fundamental in polymyalgia rheumatica (PMR) and giant cell arteritis (GCA), but carries a risk of glucocorticoid-induced adrenal insufficiency. Adrenal insufficiency can cause reluctance to stop glucocorticoid treatment after disease remission as symptoms can resemble PMR/GCA flare. We aimed to determine the prevalence of adrenal insufficiency in prednisolone-treated patients with PMR/GCA. METHODS: We included 47 patients with PMR (n = 37), GCA (n = 1) or both (n = 9), treated with prednisolone for ≥5.4 months, current dose 2.5-10 mg/day. Adrenal function was evaluated using a corticotropin (Synacthen®) stimulation test following 48 h prednisolone pause. Two years' clinical follow-up data are provided. RESULTS: Seven patients (15%) had adrenal insufficiency, 4 (11%) of the 37 patients with PMR alone, and 3 (30%) of the 10 patients with GCA. Corticotropin-stimulated P-cortisol was significantly associated with current prednisolone dose, mean daily dose the last 3 and 6 months before testing, and basal P-cortisol, but not with total dose or treatment duration. Adrenal insufficiency occurred with all current prednisolone doses (2.5-10 mg/day). Five (71%) of the glucocorticoid-insufficient patients could discontinue prednisolone treatment; two of them recovered glucocorticoid function, whereas three still needed hydrocortisone replacement 2 years later. Two patients experienced in total four acute hospital admissions with symptoms of adrenal crises. CONCLUSION: Glucocorticoid-induced adrenal insufficiency occurred in 15% of patients with PMR/GCA. Mean prednisolone dose the last 3 months and basal P-cortisol were the best and simplest predictors of adrenal function. Most of the glucocorticoid-insufficient patients could discontinue prednisolone with appropriate treatment for adrenal insufficiency.