Background: Short Synacthen test (SST) is commonly used for the assessment of the hypothalamus-pituitary-adrenal (HPA) axis after pituitary surgery. In our centre, patients are discharged on hydrocortisone on day 3 post-surgery and 9am cortisol on day 8 post-surgery is measured. Hydrocortisone is stopped if cortisol is ≥350 nmol/L on day 8. Six weeks post-surgery SST is performed and treatment adjusted. We aimed to assess the performance of day 8 morning cortisol as a predictor of the hypothalamus-pituitary-adrenal (HPA) axis status in patients who underwent transsphenoidal surgery (TSS).
Methods: We have performed a retrospective cohort analysis of 79 patients who had a TSS for non-ACTH producing pituitary tumours in Oxford between 2014 and 2017 and had 6 weeks post-surgery SST data available. Cortisol was measured using Advia-Centaur assay (Siemens). SPSS v23 was used for statistical analysis.
Results: ROC curve analysis was performed to identify a 9am cortisol value at day 8 post-TSS to predict accurately passing the SST at 6 weeks post-surgery. 9am cortisol above 433 nmol/L at day 8 had 100% specificity for predicting a normal SST at 6 week post-surgery. Our current day 8 cortisol cut-off of 350 nmol/L had 85% specificity for passing 6 week post-surgery SST, therefore 15% of patients would stop hydrocortisone inappropriately.
We also compared SST data pre- and post-TSS (n=61): 3 patients failed pre- and post-surgery SST (4.9%); 4 patients failed pre-surgery SST but subsequently passed 6 week post-surgery SST (6.5%); 10 patients had normal SST pre-TSS but failed 6 week post-surgery SST (16.4%).
Conclusions: In our cohort of patients, day 8 cortisol above 350 nmol/L led to discontinuation of steroids in 15% of cases who subsequently failed SST. Our results should be validated on an increased number of subjects aiming for a safe algorithm for post-TSS assessment of the HPA axis.