Early Morning Cortisol Level as a Predictive Factor for Long-Term Glucocorticoid Replacement After Pituitary Surgery: A Systematic Review and Meta-Analysis

Abstract

Background A reliable strategy for predicting long-term adrenal insufficiency after pituitary surgery can reduce the risk of glucocorticoid overexposure or missing patients with pituitary insufficiency. For this purpose, we aimed to assess the predictive value of early postoperative morning serum cortisol level for the detection of hypothalamic-pituitary-adrenal axis dysfunction in patients who underwent pituitary surgery. Methods A Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)–based systematic review was conducted to include articles investigating morning blood cortisol levels after pituitary surgery for lesions of the pituitary gland as a determinant for administration of long-term supplemental glucocorticoids. Bayesian statistics were used to pool the sensitivity and specificity rates. Sensitivity and specificity were also determined for each potential cortisol level on postoperative day (POD) 1 and POD 2. Results The study included 17 articles encompassing 1648 patients. Morning cortisol levels on POD 1 and POD 2 showed pooled sensitivity rates of 86.4% and 86.6% and pooled specificity rates of 73.1% and 78.2%, respectively, for predicting long-term glucocorticoid replacement after surgery. A cortisol level of 2.1 μg/dL showed the highest sensitivity rate (98.78%), and 22.5 μg/dL showed the highest specificity rate (72.5%) on POD 1. Conclusions In this review and Bayesian meta-analysis, we found that postoperative serum cortisol measurement may have high accuracy in prediction of the long-term need for glucocorticoid administration in patients who underwent pituitary surgery.