Diagnostic performance of morning serum cortisol in glucocorticoid weaning in children and adults

Arshad Muhammad Fahad; Aneeq Ahmed; Sian Beddows; Isaac Marsh; Arwa Mullamitha; John Newell-Price; Richard Ross; Charlotte Elder; Miguel Debono

doi:10.1530/endoabs.109.P285

P285

Prev Next

Section Contents Cite

Endocrine Abstracts (2025) 109 P285 | DOI: 10.1530/endoabs.109.P285

SFEBES2025 Poster Presentations Late Breaking (68 abstracts)

Diagnostic performance of morning serum cortisol in glucocorticoid weaning in children and adults

Muhammad Fahad Arshad ^1,2 , Aneeq Ahmed ^1,3 , Sian Beddows ¹ , Isaac Marsh ¹ , Arwa Mullamitha ¹ , John Newell-Price ^1,2 , Richard Ross ¹ , Charlotte Elder ^1,3 & Miguel Debono ^1,2

Author affiliations

¹University of Sheffield, Sheffield, United Kingdom. ²Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, United Kingdom. ³Sheffield Children’s NHS Foundation Trust, Sheffield, United Kingdom

Introduction: Glucocorticoid-induced suppression is the commonest cause of adrenal insufficiency (AI). The morning serum cortisol levels recommended by recent NICE guidelines to confirm or exclude adrenal insufficiency (AI) are not derived from studies in patients with AI due to glucocorticoids. The aim of this study was to identify the morning serum cortisol that predicts a normal 30-minute post-synacthen cortisol in children and adults with suspected glucocorticoid-induced AI.

Materials and Methods: A retrospective cohort study of paediatric and adult patients on long-term glucocorticoids with suspected tertiary AI undergoing a short synacthen test (SST). Main outcome: morning serum cortisol cut-offs with 95% and 99% specificity and sensitivity determined via receiver operating characteristic (ROC) curve analysis. A pass for the SST was defined as a post-synacthen 30-minute cortisol of ≥430 nmol/L using immunoassays, Vitros 5600 (Ortho Clinical Diagnostics) in paediatric cohort, and Elecsys II (Roche) in the adult cohort.

Results: 173 and 443 SSTs were included in the paediatric and adult cohorts, respectively, of which 32% and 36% were normal. The ROC curve analysis demonstrated that basal morning cortisol performed well in both cohorts with area under curve (AUC) of 0.77 (95%CI 0.70,0.85) and 0.89 (95%CI 0.85,0.92), respectively. Morning serum cortisol cut offs to predict a normal SST in children and adults were 280 and 285 nmol/L at 95% sensitivity, and 316 and 349 nmol/L at 99% sensitivity, respectively. In a longitudinal analysis carried out in adults, using the 95% cut off, 54 of 57 patients with morning serum cortisol values ≥285 nmol/L were weaned from glucocorticoids within three months.

Conclusion: A morning serum cortisol of ≥285 nmol/L means that most patients can be weaned from glucocorticoids without the need for synacthen testing.