MON-304 Prolactin as a Surrogate Marker to Predict Long Term Postoperative Hypopituitarism After Transsphenoidal Resection of Pituitary Adenomas

Abstract

Transsphenoidal surgery (TSS) is the first line treatment for pituitary adenoma. A well-known complication of TSS is hypopituitarism with a reported risk of 5-25% after resection of pituitary adenomas. A decrease in postoperative prolactin concentration was shown to be associated with postoperative hypopituitarism in a previous report. We hypothesized that in addition to clinical factors (preoperative hypofunction and adenoma size), biochemical factors (change in prolactin concentration and immediate post-operative hypofunction) can aid in predicting long term hypopituitarism as defined as ≥1 biochemically confirmed hypofunctioning pituitary axes 3 years after resection. A retrospective analysis of all patients undergoing TSS for both functioning and non-functioning pituitary adenomas at a tertiary center from January 2013 through December 2015 was performed. Prolactinomas were excluded. Of the 75 patients included, 21.3% (n=16) had at least one pituitary axis requiring replacement at 3 years post operatively. Mean age at presentation was 55 ± 16 years, 55% were female and 81% were Caucasian. Mean adenoma size was no different between normal pituitary function and hypopituitary groups (24.0 ± 11.9 mm versus 25.3 ± 10, p=0.7). Factors associated with long term hypopituitarism were older age (mean age 64 ± 4 years versus 53 ± 2 years, p = 0.02), preoperative secondary adrenal insufficiency (AM cortisol 6.4 ± 3.7 vs 12.0 ± 6.5 µg/dL; p = 0.03), preoperative secondary hypothyroidism (0.8 ± 0.2 vs 12.0 ± 6.5 ng/dL; p < 0.01), low immediate postoperative cortisol (5.3±3.1 vs 26.1±18.3 µg/d; p<0.01), and persistence of adrenal insufficiency (10.7% vs 2.7%; p<0.01) and secondary hypothyroidism (13.3% vs 5.3%; p<0.01) at 3 months. Change in prolactin concentration from preoperative to postoperative day 1-7 was not significantly different between groups (p=0.09) due to the higher variability in the hypopituitary group (median 0.2 ng/mL, IQR -0.5 - 0.8 ng/mL) compared to the normal pituitary function group (median 0.7 ng/mL, IQR 0.5-0.8 ng/mL). Adenoma size, optic chiasm and cavernous sinus involvement were not associated with long term hypopituitarism. In patients who developed postoperative hypopituitarism, there was a higher frequency of adenoma persistence or recurrence (20% vs 47%). There was a high rate of patients lost to follow up (56%). Older age, the presence of preoperative secondary adrenal insufficiency and hypothyroidism, and low day 1-7 postoperative cortisol concentration are factors that can be used to deem a patient high risk for future hypopituitarism. These patients should have close follow up with continued screening postoperatively. Contrary to prior reports, adenoma size and parasellar involvement were not associated which may be suggestive of surgical expertise. Prolactin concentrations proved not to be a good surrogate marker to predict long term hypopituitarism.