J Neurol Surg B Skull Base 2020; 81(S 01): S1-S272
DOI: 10.1055/s-0040-1702299
Oral Presentations
Georg Thieme Verlag KG Stuttgart · New York

Long-Term Hypocortisolism after Pituitary Surgery: Can POD1 Stress Response Predict Long-Term Outcome?

Pascal Lavergne
1   Thomas Jefferson University, Philadelphia, Pennsylvania, United States
,
Prachi Patel
1   Thomas Jefferson University, Philadelphia, Pennsylvania, United States
,
Chandala Chitguppi
1   Thomas Jefferson University, Philadelphia, Pennsylvania, United States
,
Tawfiq Khouri
1   Thomas Jefferson University, Philadelphia, Pennsylvania, United States
,
Mindy Rabinowitz
1   Thomas Jefferson University, Philadelphia, Pennsylvania, United States
,
Gurston G. Nyquist
1   Thomas Jefferson University, Philadelphia, Pennsylvania, United States
,
Marc R. Rosen
1   Thomas Jefferson University, Philadelphia, Pennsylvania, United States
,
Christopher J. Farrell
1   Thomas Jefferson University, Philadelphia, Pennsylvania, United States
,
James J. Evans
1   Thomas Jefferson University, Philadelphia, Pennsylvania, United States
› Author Affiliations
Further Information

Publication History

Publication Date:
05 February 2020 (online)

 

Introduction: Long-term function of the hypothalamic-pituitary-adrenal (HPA) axis is an important metric following pituitary surgery. Although universal stress dose and postoperative steroids have been a common practice for pituitary surgery, many centers now tailor postoperative steroid therapy based on patients’ individual pre- and early postoperative cortisol values. Previous studies have identified the predictive value of cortisol levels drawn on postoperative days (POD) 1, 2, and 6 as useful tool to predict immediate and long-term hypocortisolism. While a low value should prompt steroid replacement, a normal value early in the postoperative course has not traditionally been viewed as a strong predictor of long-term normal function. Cortisol levels drawn on POD 1 can show high values which are not seen on POD2, likely representing a normal stress response to surgery. The predictive value of POD 1 cortisol levels on long-term HPA axis function remains to be elucidated.

Methods: A single-institution retrospective review of patients who underwent endoscopic transnasal transsphenoidal surgery for a pituitary adenoma between 2014 and 2017 (ongoing recruitment) was performed. Patients with preoperative hypocortisolism or insufficient data in the medical record were excluded from the study. Baseline demographic data, tumor characteristics, and endocrine results were collected. Chi-squared, Kendall’s tau and Spearman’s rho, and t-tests were used to assess relationships between categorical and continuous variables and long-term hypocortisolism. Long-term hypocortisolism was defined as the need for exogenous steroids more than 6 months after surgery.

Results: One hundred and twenty-six nonsecretory pituitary adenoma patients were included in this review. Twenty-seven of the 126 (21%) patients needed steroid replacement 6 months or more after surgery. Male patients were more likely to require long-term replacement with an odds ratio of 2.065 (1.005–4.241). POD 1, POD 2, and PO 3 months’ cortisol values were highly correlated with the need for long-term steroids. POD 1 morning cortisol had the highest correlation with long-term hypocortisolism with a hazard ratio (HR) of 17.996 (7.558–28.435) compared with POD 2 values (HR = 8.980 [1.910–16.051]). Mean POD 1 cortisol value for patients without long-term hypocortisolism was significantly higher compared with the patient with long-term hypocortisolism (37.32 [11.77–62.88] vs. 19.33 [0.99–37.66] ng/mL). Tumor size and recurrent tumor were not associated with long-term hypocortisolism.

Conclusion: POD 1 cortisol is a strong predictor of long-term hypocortisolism. Although a combination of POD 1 and POD 2 has been advocated to accurately predict the need for steroid replacement therapy, our results suggest that a POD 1 level alone can be sufficient. Stress response on POD 1 with high morning cortisol value seems to be correlated with long-term normal function of the HPA axis. These results suggest that the evaluation of the cortisol level alone should not be a criterion to keep patients in the hospital past the first POD. Outpatient cortisol monitoring will still be needed to identify the low percentage of patients that may develop long-term hypocortisolism with a normal POD1 value. This study also emphasizes the previous finding that stress dose steroids are not uniformly required postoperatively.