Abstract
Hormonal insufficiency of 1 or more pituitary axes can appear after pituitary surgery. Adrenal axis impairment after surgery can lead to serious consequences if not identified and treated.
Assess early and late postoperative adrenal insufficiency (AI) and identify the risk factors predicting their occurrence after endoscopic transsphenoidal resection of pituitary adenomas.
Retrospective review identified 176 pituitary adenomas resected using an endoscopic transsphenoidal approach. Patients taking steroids preoperatively, Cushing disease patients, and patients with incomplete records were excluded. Sixty-nine patients were excluded according to our exclusion criteria.
The study group thus included 107 patients (total of 111 operations). The median age was 50 years (range, 18–89 years). The median duration of follow-up was 30 months (range, 6–74 months). Eighty-three patients (74.7%) had macroadenomas, and 89 (59.3%) had nonfunctional adenomas. Of the 111 procedures, 61 patients (55%) had early AI. Of the 61 patients, 48 patients (79%) were not taking steroids in long-term follow-up, and only 13 (21%) required long-term replacement. Sixteen of the patients undergoing 111 procedures (14.4%) had AI on long-term follow-up. Of those 16 patients, 13 were already taking steroids and 3 had new diagnoses of AI. Age, male gender, and cerebrospinal fluid (CSF) leaks were associated with persistent postoperative AI (P = 0.018, P = 0.001, P = 0.007, respectively).
Hypothalamic pituitary adrenal axis insufficiency is common after endoscopic transsphenoidal surgery. Male gender, age greater than 50 years, visual impairment, and intraoperative CSF leak were correlated with late postoperative AI. More than two thirds of patients in whom early AI developed did not require steroids in the long term.
•Of 111 endoscopic transsphenoidal surgery for pituitary adenoma, 61 patients (55%) had early adrenal insufficiency (AI). Of the 61 patients, 48 patients (79%) were not using steroids at long-term follow-up and only 13 (21%) required long-term replacement.•16/111 patients (14.4%) had AI at long-term follow-up. Of those 16 patients, 13 were already using steroids and 3 had new diagnoses of AI.•Male gender, age older than 50 years, visual impairment, and intraoperative cerebrospinal fluid leak were correlated with late postoperative AI.•More than two thirds of patients in whom early AI developed did not require steroids in the long term.