Background: Patients with severe asthma often need oral corticosteroid (OCS) treatment for uncontrolled symptoms and exacerbations. However, OCS use is associated with side effects among which adrenal insufficiency is one of the most insidious and dangerous.

Aims and objectives: To measure adrenal function by morning cortisol in a population of severe asthmatics and to correlate with OCS treatment.

Methods: A total of 71 consecutive severe asthmatics have been measured for serum morning cortisol and classified in three groups: confirmed adrenal insufficiency (AI) (?3?g/dl), uncertain diagnosis (3>?g/dl ?18?g/dl), normal (>18?g/dl). Patients required to be OCS dependent defined as the need of at least 3 OCS bursts (OCS-B) or chronic OCS intake (OCS-C) in the last year or before biologic treatment. Cumulative prednisone dose/year (PD/Y) and clinical characteristics were correlated to adrenal function.

Results: The majority of patients were female (63.4%), 42,7% needing chronic OCS for a mean of 550 days. The remaining 57,3% were used to assume OCS bursts. Mean cortisol value was 9.9 ?g/dl [range 0.1-24], confirmed AI was found in 18.3%, uncertain diagnosis in 77.5% and normal in 4.2%. Substitutive therapy was started in 84.6% and 21.8% of confirmed AI and uncertain diagnosis group respectively. PD/Y was significantly higher in the OCS-C compared to OCS-B (p<0.0001), but not between confirmed AI and uncertain diagnosis group (2130±855mg and 2469±603mg, p=0,963). The prevalence of patients with confirmed AI is not different between OCS-B and OCS-C (p=0,217)

Conclusions: Adrenal insufficiency is very common in corticosteroid dependent asthmatics, even in OCS burst, and often needing substitutive treatment