Endocrine Abstracts

Background: Glucagon (GST) is used as an alternative to insulin (ITT) to diagnose GH deficiency (GHD) and adrenal insufficiency (ACTHd). However the peak cortisol response to diagnose adrenal insufficiency varies (550, and 500 nmol/l), has been extrapolated from adults undergoing intraoperative stress and not fully validated.

Aim: To determine the peak cortisol ‘cut off’ level which most accurately predicts clinically significant adrenal insufficiency requiring therapy in children undergoing GST for suspected hypopituitarism or short stature.

Method: 140 records of all patients undergoing GST at our tertiary centre between January 2011 and December 2011 were retrospectively assessed. Patients on steroid therapy (n=1) or with incomplete test results (n=2) were excluded. All 6 cortisol components of the 3 h test results were collected. The sensitivity and specificity of different cut-off peak cortisol levels in accurately diagnosing adrenal insufficiency were calculated.

Results: Of 137 patients, aged (median) 8.7 (range 0.94–19.17) years, 112 (82%) achieved the peak cortisol response between 150 and 180 min but in 14 (10%) this was achieved at baseline. 29 (21%) and 25 (18%) had peak cortisol responses of <550 and <500 nmol/l respectively. Of these only 7 (24%) clinically warranted further investigation and 4 (14%) were then confirmed adrenally insufficient and received treatment. The rest 25 (86%) remained well without hydrocortisone replacement 6 months later.

Conclusion: Lower cut-off peak cortisol of <450 nmol/l increases the specificity without altering the sensitivity in diagnosing childhood adrenal insufficiency using a GST and may be more reflective of adrenal capacity than higher levels (550, and 500 nmol/l).