The gold standard for assessing the hypothalamic-pituitary-adrenal axis is an insulin tolerance test (ITT), which examines the cortisol response to insulin-induced hypoglycaemia. A peak cortisol cut-off of 550nmol/l is regarded as the minimum response for healthy individuals during an ITT. However, this cut-off is based on outdated assays, and risks misdiagnosis of adrenal insufficiency (AI).

In this study, Lazarus et al. undertake a retrospective analysis of 300 ITT cortisol responses using the Abbott Architect and Alinity analyser platforms in patients with suspected AI between 2010 and 2022, at a tertiary London centre. Based on the institution’s current peak cut-off value of 500nmol, receiver operating characteristic analysis identified a 100.0% sensitivity and 43.6% specificity in correctly diagnosing AI. Using a lower cortisol threshold value of 416nmol/l on the Abbott analyser platform maintained a sensitivity of 100.0% and improved the specificity to 86.7%.

The authors therefore provide evidence to update and lower the Abbott analyser cortisol threshold from 500 to 416nmol/l. Use of this threshold avoids misdiagnosis of AI and subsequent unnecessary glucocorticoid replacement therapy in otherwise healthy patients. 

Read the full article in Clinical Endocrinology https://doi.org/10.1111/cen.15074