Peri-Operative Glucocorticoids Can be Withheld Safely in Patients with an Intact Hypothalamic-Pituitary-Adrenal Axis Undergoing Transsphenoidal Surgery of a Pituitary Adenoma or Cyst

Presentation Number: SAT-0652
Date of Presentation: June 21st, 2014

Julie M Silverstein*1, Bithika M Thompson2, Jie Zheng2, Stacy Hurst2, Albert H Kim2, Ralph G Dacey2, Keith M Rich2, Gregory J Zipfel2, Michael R Chicoine2 and Clay F Semenkovich2
1Washington University School of Medicine, St Louis, MO, 2Washington University School of Medicine, St. Louis, MO


There is no consensus regarding optimal peri-operative steroid management of patients with pituitary adenomas or cysts undergoing pituitary surgery.  Patients undergoing pituitary surgery generally receive “stress” doses of glucocorticoids around the time of surgery to prevent adrenal insufficiency, but compelling data to support the routine use of peri-operative steroids are unavailable. Given the pleiotropic adverse effects of steroids, appropriately avoiding glucocorticoid use could decrease peri-operative complications. We conducted a pilot prospective study to test the hypothesis that withholding glucocorticoids in patients with an intact hypothalamic-pituitary-adrenal (HPA) axis undergoing transsphenoidal surgery is safe.  Patients (including those with non-functioning adenomas, acromegaly, TSH-secreting adenomas, or prolactinomas) with an intact HPA axis scheduled to undergo transsphenoidal resection of a pituitary adenoma or cyst were randomized either to receive IV hydrocortisone at the time of anesthesia induction and then dexamethasone 0.5mg IV q 6 x 4 doses (STER group, n= 11) or to undergo surgery without peri-operative steroids (NOSTER group, n= 8).  For all subjects, an 8 am cortisol level was determined either on post-operative day one (NOSTER group) or on post-operative days two or three, at least 24 hours after the last dose of dexamethasone (STER group).  Patients whose 8 am cortisol was < 15 mcg/dl were treated with prednisone 5mg daily.  All patients underwent a cosyntropin stimulation test four to six weeks after surgery and prednisone was discontinued if testing was consistent with an intact HPA axis.  Data regarding length of hospital stay, incidence of diabetes insipidus (DI), hyperglycemia, incidence of adrenal insufficiency (AI), development of delayed hyponatremia, and complications were collected.  There was no difference in post-operative glucose levels in non-diabetic patients (121 +/-  5 mg/dl vs. 128 +/- 17 mg/dl, p= 0.15), in the development of transient  DI (54.6% vs. 42.9%, p= 1.0), in the development of delayed hyponatremia (9.1% vs. 28.6%, p= 0.53), in the number of patients discharged on glucocorticoids (27.3% vs. 14.3%, p= 1.0) or in the number of patients who developed permanent AI (9.1% vs. 0%, p=1.0) in the STER and NOSTER  groups respectively.  One patient in the NOSTER group died of post-operative meningitis and there were no deaths or serious complications in the STER group. There was a trend towards an increase in length of hospital stay in the STER group compared to the NOSTER group (4.36 days vs. 3.71 days, p= 0.418).  While this study is limited by small sample sizes that increase the possibility of a type II error, these results provide data suitable for designing a larger study to confirm that withholding glucocorticoids in patients with an intact HPA axis undergoing transsphenoidal resection of pituitary adenomas or cysts may be beneficial.


Disclosure: JMS: Clinician, Pfizer, Inc.. CFS: Researcher, Merck & Co., Clinician, Merck & Co.. Nothing to Disclose: BMT, JZ, SH, AHK, RGD, KMR, GJZ, MRC