Aim: Shortage of tetracosactide in the UK has put pressure on clinicians to reduce the number of SSTs ordered. Our aim was to establish if a lower baseline cortisol level from the currently accepted cut-off could predict a positive SST outcome. This would aid our Endocrine Department in rationalising use of tetracosactide.
Methods: A retrospective observational study of indications and results of SSTs performed in non-critically ill general medical patients who were referred to the Endocrine OPD. 290 SST tests were done between October 2014 to January 2019 of which, 103 were excluded as baseline cortisol was taken after 1000 h. Plasma cortisol was measured at baseline and 30 min following administration of tetracosactide 250 (g. The locally validated assay specific cut-off for passed SST was ≥550 nmol/l (Roche Cobas e601 analyser). Receiver operating Characteristic (ROC) curve was generated to determine the predictive value of basal cortisol for a passed SST.
Results: Indications for referral included; establishing a new diagnosis (52%), assessing adrenal axis in patients with other endocrine disorders (37%), assessing safety of stopping long term steroids (10%) and undocumented (1%). 187 SSTs were analysed of which 139 (74%) passed and 48 (26%) failed. ROC curve analysis identified that basal cortisol of <130 nmol/l predicted failure with 100% sensitivity and ≥400 nmol/l provided 100% specificity to pass SST. Using the minimum ROC distance criterion, basal cortisol value of ≥300 nmol/l was identified to predict passing SST with 72% sensitivity and 77% specificity.
Discussion: Our findings suggest that baseline cortisol ≥400 nmol/l could have prevented SSTs as it predicts failed SST. In this cohort 50 SSTs could have been potentially avoided with significant cost saving implications. However, we continue to recommend SST as further studies are needed to validate this baseline cortisol cut-off.