Background: Up to 50% of children treated with inhaled corticosteroids (ICS) have biochemical evidence of adrenal insufficiency (AI). Episodes of adrenal crisis (AC) are extremely rare.
To address the discordance between the prevalence of biochemical AI and AC, we re-examined the biochemical definition of AI during ICS therapy. We then investigated the utility of early morning salivary cortisol (EMSC) and cortisone (EMSCn) for the identification of patients with AI using standard and revised diagnostic criteria.
Patients and methods: Cases of AC during ICS therapy in childhood were identified on PubMed. Data from these cases were used to inform a revised definition of AI on the LDSST.
Subjects (160 M), median age 10.0 (5.115.2) years collected saliva samples for 3 consecutive days and underwent the LDSST (Synacthen 500 ng/1.73 m2) on D3. The relationship between EMSC and EMSCn and peak cortisol (PC) on the LDSST and was examined using standard and revised definitions of AI.
Results: Fifty cases of AC were identified in PubMed. Dynamic function tests were reported in 34 subjects: PC (mean) was 44 (maximum 387) nmol/l on standard dose short Synacthen stimulation test (n=29), 127 and 107 nmol/l on LDSST (n=2) and 150 (312) nmol/l on glucagon stimulation test (n=3). The modified definition of AI was defined as PC <350 nmol/l on LDSST.
In our cohort, PC was <500 nmol/l in 101/269 subjects (37.5%) and <350 nmol/l in 12/269 subjects (4.5%) on the LDSST. EMSC and EMSCn had no predictive value for PC <500 nmol/l. EMSCn cut-off value of 12.5 nmol/l gave a negative predictive value of 99.2% and positive predictive value of 30.1% for PC <350 nmol/l.
Discussion: Screening subjects using EMSCn, followed by a LDSST using PC <350 nmol/l define AI, may enable a meaningful and practical method of identifying subjects at greatest risk of AC during ICS therapy.
13 Nov 2013
British Society for Paediatric Endocrinology and Diabetes