Predicting Recovery of the Hypothalamic-Pituitary-Adrenal Axis After Prolonged Glucocorticoid Use

Sii Hoe Leong, MBBS; Shubash Shander, MBBS, MPH, MBA; Jeyakantha Ratnasingam, MD, MMed


Endocr Pract. 2018;24(1):14-20. 

In This Article

Abstract and Introduction


Objective: Prolonged exposure to glucocorticoids lead to hypothalamic-pituitary-adrenal (HPA) axis suppression that recovers after cessation of treatment. We aimed to identify the predictive factors for HPA axis recovery after prolonged glucocorticoid use.

Methods: Retrospective review of patients who had undergone first short Synacthen test (SST) to assess HPA axis recovery after prolonged use of glucocorticoids.

Results: A total of 61% (20/33) of patients had adequate SST response at a median time of 2 years after diagnosis of adrenal insufficiency. Those who had adequate response during SST had higher ambulatory early morning cortisol (P<.01), shorter duration of exposure to glucocorticoids (P = .01), and lower final cumulative hydrocortisone replacement dose (P = .03). Age, gender, body mass index, indications for glucocorticoid use, and basal adrenocorticotropic hormone levels were not predictive of HPA axis recovery. On multivariate analysis, ambulatory early morning cortisol was the only independent predictor of adequate SST response (odds ratio, 1.02; 95% confidence interval, 1.01 to 1.04; P = .02). Using receiver operating characteristic curve analysis, ambulatory early morning cortisol of 8.8 μg/dL predicted a positive SST response with a sensitivity of 70% and specificity of 93%.

Conclusion: Early morning ambulatory cortisol could be used to decide on timely SST in order to prevent complications from unnecessary replacement with glucocorticoids.


Glucocorticoids have commonly been used for decades to treat a myriad of conditions that are autoimmune or inflammatory in origin.[1] In Asia, glucocorticoids are also widely prescribed factitiously by traditional healers to provide symptomatic relief from many common ailments, including joint pain, rash, allergy, or even lethargy. These remedies that include Chinese traditional medicine and other homeopathic treatments have been reported to contain excess amounts of undeclared corticosteroids, namely dexamethasone, in addition to other adulterants.[2–4]

Chronic use of glucocorticoids leads to undesirable metabolic complications, including weight gain, diabetes, hypertension, osteoporosis, and more importantly, steroid withdrawal symptoms and hypothalamic-pituitary-adrenal (HPA) axis suppression if the glucocorticoids are discontinued abruptly.[5,6] Therefore, timely initiation of physiologic doses of glucocorticoid replacement until recovery of the adrenal function is crucial.[7] Recovery of the HPA axis occurs between 1 and 3 years of cessation of glucocorticoids, but this time span is not well established.[8,9] Although the gold-standard test to detect HPA axis recovery would be the insulin tolerance test (ITT), this test is cumbersome to perform. The short Synacthen test (SST), which is easily performed at any time of the day, has been shown to be highly correlated with the ITT and could be used as a reliable test to assess HPA axis recovery.[10–12] Failure to detect recovery of the HPA axis leads to unnecessary administration of excess glucocorticoids in a person who is replete with endogenous glucocorticoids, leading to adverse cardiometabolic complications of excess glucocorticoids.

To date, there is no clear guidance for the managing clinician to predict recovery of the HPA axis in order to decide on the best timing for HPA axis testing.[13,14] There are also sparse data on recovery of the HPA axis, especially after use of traditional medicines. As this is a commonly encountered cause of HPA axis suppression in our clinical practice, we decided to embark on this study to determine the predictive factors for HPA axis recovery after prolonged glucocorticoid use.